diff --git a/blueprints/ControlNet (Z-Image-Turbo).json b/blueprints/ControlNet (Z-Image-Turbo).json new file mode 100644 index 000000000..fbec95a97 --- /dev/null +++ b/blueprints/ControlNet (Z-Image-Turbo).json @@ -0,0 +1,1412 @@ +{ + "revision": 0, + "last_node_id": 85, + "last_link_id": 0, + "nodes": [ + { + "id": 85, + "type": "d2e76ecf-6e84-4b8c-8913-48efc09ec1c4", + "pos": [ + 440, + 1220 + ], + "size": [ + 480, + 0 + ], + "flags": {}, + "order": 6, + "mode": 0, + "inputs": [ + { + "label": "control_image", + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": null + }, + { + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": null + }, + { + "name": "seed", + "type": "INT", + "widget": { + "name": "seed" + }, + "link": null + }, + { + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": null + }, + { + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": null + }, + { + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": null + }, + { + "label": "patch_model", + "name": "name", + "type": "COMBO", + "widget": { + "name": "name" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [] + } + ], + "title": "ControlNet (Z-Image-Turbo)", + "properties": { + "proxyWidgets": [ + [ + "83", + "text" + ], + [ + "79", + "seed" + ], + [ + "74", + "unet_name" + ], + [ + "73", + "clip_name" + ], + [ + "75", + "vae_name" + ], + [ + "76", + "name" + ], + [ + "79", + "control_after_generate" + ] + ], + "cnr_id": "comfy-core", + "ver": "0.18.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [] + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "d2e76ecf-6e84-4b8c-8913-48efc09ec1c4", + "version": 1, + "state": { + "lastGroupId": 9, + "lastNodeId": 85, + "lastLinkId": 87, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "ControlNet (Z-Image-Turbo)", + "inputNode": { + "id": -10, + "bounding": [ + -500, + 620, + 120, + 180 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + 1390, + 1100, + 120, + 60 + ] + }, + "inputs": [ + { + "id": "fbbb968e-d3cf-40e4-b3ce-7abb074e5bd8", + "name": "image", + "type": "IMAGE", + "linkIds": [ + 65, + 80 + ], + "localized_name": "image", + "label": "control_image", + "pos": [ + -400, + 640 + ] + }, + { + "id": "c1b19877-5417-4580-aea1-44439c70c1dd", + "name": "text", + "type": "STRING", + "linkIds": [ + 81 + ], + "pos": [ + -400, + 660 + ] + }, + { + "id": "b5671515-bc7a-4be5-b1e7-d4f0f68907d6", + "name": "seed", + "type": "INT", + "linkIds": [ + 83 + ], + "pos": [ + -400, + 680 + ] + }, + { + "id": "2838be23-8034-4f16-87a5-d29d790e8391", + "name": "unet_name", + "type": "COMBO", + "linkIds": [ + 84 + ], + "pos": [ + -400, + 700 + ] + }, + { + "id": "8a6643b5-8f78-41ff-bbc6-e87b95459706", + "name": "clip_name", + "type": "COMBO", + "linkIds": [ + 85 + ], + "pos": [ + -400, + 720 + ] + }, + { + "id": "b103dc94-8ca7-456b-a809-414d7e341a1b", + "name": "vae_name", + "type": "COMBO", + "linkIds": [ + 86 + ], + "pos": [ + -400, + 740 + ] + }, + { + "id": "4a7d65af-f0fd-4a5c-832a-bdc0d15b1f30", + "name": "name", + "type": "COMBO", + "linkIds": [ + 87 + ], + "label": "patch_model", + "pos": [ + -400, + 760 + ] + } + ], + "outputs": [ + { + "id": "ccb7fa39-4a3d-4eb2-8fd2-91d08fad9570", + "name": "IMAGE", + "type": "IMAGE", + "linkIds": [ + 45 + ], + "localized_name": "IMAGE", + "pos": [ + 1410, + 1120 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 73, + "type": "CLIPLoader", + "pos": [ + 20, + 500 + ], + "size": [ + 270, + 150 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "localized_name": "clip_name", + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": 85 + }, + { + "localized_name": "type", + "name": "type", + "type": "COMBO", + "widget": { + "name": "type" + }, + "link": null + }, + { + "localized_name": "device", + "name": "device", + "shape": 7, + "type": "COMBO", + "widget": { + "name": "device" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [ + 44 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.73", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "CLIPLoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "qwen_3_4b.safetensors", + "url": "https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/text_encoders/qwen_3_4b.safetensors", + "directory": "text_encoders" + } + ] + }, + "widgets_values": [ + "qwen_3_4b.safetensors", + "lumina2", + "default" + ] + }, + { + "id": 74, + "type": "UNETLoader", + "pos": [ + 20, + 320 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "unet_name", + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": 84 + }, + { + "localized_name": "weight_dtype", + "name": "weight_dtype", + "type": "COMBO", + "widget": { + "name": "weight_dtype" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 79 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.73", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "UNETLoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "z_image_turbo_bf16.safetensors", + "url": "https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/diffusion_models/z_image_turbo_bf16.safetensors", + "directory": "diffusion_models" + } + ] + }, + "widgets_values": [ + "z_image_turbo_bf16.safetensors", + "default" + ] + }, + { + "id": 75, + "type": "VAELoader", + "pos": [ + 20, + 760 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "vae_name", + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": 86 + } + ], + "outputs": [ + { + "localized_name": "VAE", + "name": "VAE", + "type": "VAE", + "links": [ + 39, + 70 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.73", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "VAELoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "ae.safetensors", + "url": "https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/vae/ae.safetensors", + "directory": "vae" + } + ] + }, + "widgets_values": [ + "ae.safetensors" + ] + }, + { + "id": 76, + "type": "ModelPatchLoader", + "pos": [ + 20, + 940 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "name", + "name": "name", + "type": "COMBO", + "widget": { + "name": "name" + }, + "link": 87 + } + ], + "outputs": [ + { + "localized_name": "MODEL_PATCH", + "name": "MODEL_PATCH", + "type": "MODEL_PATCH", + "links": [ + 74 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.51", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ModelPatchLoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "Z-Image-Turbo-Fun-Controlnet-Union.safetensors", + "url": "https://huggingface.co/alibaba-pai/Z-Image-Turbo-Fun-Controlnet-Union/resolve/main/Z-Image-Turbo-Fun-Controlnet-Union.safetensors", + "directory": "model_patches" + } + ] + }, + "widgets_values": [ + "Z-Image-Turbo-Fun-Controlnet-Union.safetensors" + ] + }, + { + "id": 77, + "type": "VAEDecode", + "pos": [ + 940, + 1100 + ], + "size": [ + 230, + 100 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [ + { + "localized_name": "samples", + "name": "samples", + "type": "LATENT", + "link": 38 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 39 + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "slot_index": 0, + "links": [ + 45 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.64", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "VAEDecode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 78, + "type": "ModelSamplingAuraFlow", + "pos": [ + 910, + 270 + ], + "size": [ + 290, + 110 + ], + "flags": {}, + "order": 5, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 69 + }, + { + "localized_name": "shift", + "name": "shift", + "type": "FLOAT", + "widget": { + "name": "shift" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "slot_index": 0, + "links": [ + 40 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.64", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ModelSamplingAuraFlow", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 3 + ] + }, + { + "id": 79, + "type": "KSampler", + "pos": [ + 910, + 430 + ], + "size": [ + 300, + 570 + ], + "flags": {}, + "order": 6, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 40 + }, + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "link": 41 + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "link": 42 + }, + { + "localized_name": "latent_image", + "name": "latent_image", + "type": "LATENT", + "link": 78 + }, + { + "localized_name": "seed", + "name": "seed", + "type": "INT", + "widget": { + "name": "seed" + }, + "link": 83 + }, + { + "localized_name": "steps", + "name": "steps", + "type": "INT", + "widget": { + "name": "steps" + }, + "link": null + }, + { + "localized_name": "cfg", + "name": "cfg", + "type": "FLOAT", + "widget": { + "name": "cfg" + }, + "link": null + }, + { + "localized_name": "sampler_name", + "name": "sampler_name", + "type": "COMBO", + "widget": { + "name": "sampler_name" + }, + "link": null + }, + { + "localized_name": "scheduler", + "name": "scheduler", + "type": "COMBO", + "widget": { + "name": "scheduler" + }, + "link": null + }, + { + "localized_name": "denoise", + "name": "denoise", + "type": "FLOAT", + "widget": { + "name": "denoise" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "slot_index": 0, + "links": [ + 38 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.64", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "KSampler", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 729703840979498, + "randomize", + 8, + 1, + "res_multistep", + "simple", + 1 + ] + }, + { + "id": 80, + "type": "ConditioningZeroOut", + "pos": [ + 610, + 830 + ], + "size": [ + 230, + 80 + ], + "flags": { + "collapsed": true + }, + "order": 7, + "mode": 0, + "inputs": [ + { + "localized_name": "conditioning", + "name": "conditioning", + "type": "CONDITIONING", + "link": 36 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 42 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.73", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ConditioningZeroOut", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 81, + "type": "QwenImageDiffsynthControlnet", + "pos": [ + 490, + 970 + ], + "size": [ + 290, + 200 + ], + "flags": {}, + "order": 8, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 79 + }, + { + "localized_name": "model_patch", + "name": "model_patch", + "type": "MODEL_PATCH", + "link": 74 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 70 + }, + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 65 + }, + { + "localized_name": "mask", + "name": "mask", + "shape": 7, + "type": "MASK", + "link": null + }, + { + "localized_name": "strength", + "name": "strength", + "type": "FLOAT", + "widget": { + "name": "strength" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 69 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.76", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "QwenImageDiffsynthControlnet", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 1 + ] + }, + { + "id": 82, + "type": "EmptySD3LatentImage", + "pos": [ + 40, + 1200 + ], + "size": [ + 260, + 170 + ], + "flags": {}, + "order": 9, + "mode": 0, + "inputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": 76 + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": 77 + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "widget": { + "name": "batch_size" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "slot_index": 0, + "links": [ + 78 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.64", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "EmptySD3LatentImage", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 1024, + 1024, + 1 + ] + }, + { + "id": 83, + "type": "CLIPTextEncode", + "pos": [ + 430, + 310 + ], + "size": [ + 400, + 440 + ], + "flags": {}, + "order": 10, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 44 + }, + { + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": 81 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 36, + 41 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.73", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "CLIPTextEncode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "" + ], + "color": "#232", + "bgcolor": "#353" + }, + { + "id": 84, + "type": "GetImageSize", + "pos": [ + 50, + 1410 + ], + "size": [ + 230, + 120 + ], + "flags": { + "collapsed": true + }, + "order": 11, + "mode": 0, + "inputs": [ + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 80 + } + ], + "outputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "links": [ + 76 + ] + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "links": [ + 77 + ] + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "links": null + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.76", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "GetImageSize", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + } + ], + "groups": [ + { + "id": 3, + "title": "Prompt", + "bounding": [ + 410, + 230, + 440, + 630 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 4, + "title": "Model", + "bounding": [ + -50, + 230, + 430, + 840 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 8, + "title": "Apple ControlNet", + "bounding": [ + 410, + 890, + 440, + 330 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 9, + "title": "Image Size", + "bounding": [ + -50, + 1100, + 430, + 350 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + } + ], + "links": [ + { + "id": 38, + "origin_id": 79, + "origin_slot": 0, + "target_id": 77, + "target_slot": 0, + "type": "LATENT" + }, + { + "id": 39, + "origin_id": 75, + "origin_slot": 0, + "target_id": 77, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 69, + "origin_id": 81, + "origin_slot": 0, + "target_id": 78, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 40, + "origin_id": 78, + "origin_slot": 0, + "target_id": 79, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 41, + "origin_id": 83, + "origin_slot": 0, + "target_id": 79, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 42, + "origin_id": 80, + "origin_slot": 0, + "target_id": 79, + "target_slot": 2, + "type": "CONDITIONING" + }, + { + "id": 78, + "origin_id": 82, + "origin_slot": 0, + "target_id": 79, + "target_slot": 3, + "type": "LATENT" + }, + { + "id": 36, + "origin_id": 83, + "origin_slot": 0, + "target_id": 80, + "target_slot": 0, + "type": "CONDITIONING" + }, + { + "id": 79, + "origin_id": 74, + "origin_slot": 0, + "target_id": 81, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 74, + "origin_id": 76, + "origin_slot": 0, + "target_id": 81, + "target_slot": 1, + "type": "MODEL_PATCH" + }, + { + "id": 70, + "origin_id": 75, + "origin_slot": 0, + "target_id": 81, + "target_slot": 2, + "type": "VAE" + }, + { + "id": 76, + "origin_id": 84, + "origin_slot": 0, + "target_id": 82, + "target_slot": 0, + "type": "INT" + }, + { + "id": 77, + "origin_id": 84, + "origin_slot": 1, + "target_id": 82, + "target_slot": 1, + "type": "INT" + }, + { + "id": 44, + "origin_id": 73, + "origin_slot": 0, + "target_id": 83, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 65, + "origin_id": -10, + "origin_slot": 0, + "target_id": 81, + "target_slot": 3, + "type": "IMAGE" + }, + { + "id": 80, + "origin_id": -10, + "origin_slot": 0, + "target_id": 84, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 45, + "origin_id": 77, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 81, + "origin_id": -10, + "origin_slot": 1, + "target_id": 83, + "target_slot": 1, + "type": "STRING" + }, + { + "id": 83, + "origin_id": -10, + "origin_slot": 2, + "target_id": 79, + "target_slot": 4, + "type": "INT" + }, + { + "id": 84, + "origin_id": -10, + "origin_slot": 3, + "target_id": 74, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 85, + "origin_id": -10, + "origin_slot": 4, + "target_id": 73, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 86, + "origin_id": -10, + "origin_slot": 5, + "target_id": 75, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 87, + "origin_id": -10, + "origin_slot": 6, + "target_id": 76, + "target_slot": 0, + "type": "COMBO" + } + ], + "extra": { + "workflowRendererVersion": "LG" + }, + "category": "Image generation and editing/ControlNet", + "description": "Generates images from a text prompt and ControlNet conditioning (e.g. depth, canny) using Z-Image-Turbo." + } + ] + }, + "extra": { + "ue_links": [] + } +} \ No newline at end of file diff --git a/blueprints/Depth to Video (ltx 2.0).json b/blueprints/Depth to Video (ltx 2.0).json index bb28695a2..bd51e4476 100644 --- a/blueprints/Depth to Video (ltx 2.0).json +++ b/blueprints/Depth to Video (ltx 2.0).json @@ -4234,7 +4234,7 @@ "workflowRendererVersion": "LG" }, "category": "Video generation and editing/Depth to video", - "description": "Generates video from depth maps using LTX-2, with optional synchronized audio." + "description": "Generates depth-controlled video with LTX-2: motion and structure follow a depth-reference video alongside text prompting, optional first-frame image conditioning, with optional synchronized audio." }, { "id": "38b60539-50a7-42f9-a5fe-bdeca26272e2", diff --git a/blueprints/First-Last-Frame to Video.json b/blueprints/First-Last-Frame to Video.json new file mode 100644 index 000000000..84dfafbcd --- /dev/null +++ b/blueprints/First-Last-Frame to Video.json @@ -0,0 +1,3361 @@ +{ + "revision": 0, + "last_node_id": 227, + "last_link_id": 0, + "nodes": [ + { + "id": 227, + "type": "283e4561-61a2-4538-b960-265736eb041f", + "pos": [ + 620, + 3140 + ], + "size": [ + 540, + 0 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "label": "first_frame", + "localized_name": "input", + "name": "input", + "type": "IMAGE,MASK", + "link": null + }, + { + "label": "last_frame", + "localized_name": "input_1", + "name": "input_1", + "type": "IMAGE,MASK", + "link": null + }, + { + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": null + }, + { + "label": "width", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": null + }, + { + "label": "height", + "name": "value_1", + "type": "INT", + "widget": { + "name": "value_1" + }, + "link": null + }, + { + "label": "duration", + "name": "value_2", + "type": "INT", + "widget": { + "name": "value_2" + }, + "link": null + }, + { + "label": "fps", + "name": "value_3", + "type": "INT", + "widget": { + "name": "value_3" + }, + "link": null + }, + { + "name": "noise_seed", + "type": "INT", + "widget": { + "name": "noise_seed" + }, + "link": null + }, + { + "label": "ckpt_name", + "name": "ckpt_name_1", + "type": "COMBO", + "widget": { + "name": "ckpt_name_1" + }, + "link": null + }, + { + "name": "text_encoder", + "type": "COMBO", + "widget": { + "name": "text_encoder" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "VIDEO", + "name": "VIDEO", + "type": "VIDEO", + "links": [] + } + ], + "title": "First-Last-Frame to Video", + "properties": { + "proxyWidgets": [ + [ + "222", + "text" + ], + [ + "215", + "value" + ], + [ + "216", + "value" + ], + [ + "198", + "value" + ], + [ + "205", + "value" + ], + [ + "196", + "noise_seed" + ], + [ + "224", + "ckpt_name" + ], + [ + "225", + "text_encoder" + ] + ], + "cnr_id": "comfy-core", + "ver": "0.18.1", + "ue_properties": { + "widget_ue_connectable": {}, + "input_ue_unconnectable": {}, + "version": "7.7" + } + }, + "widgets_values": [] + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "283e4561-61a2-4538-b960-265736eb041f", + "version": 1, + "state": { + "lastGroupId": 22, + "lastNodeId": 227, + "lastLinkId": 276, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "First-Last-Frame to Video", + "inputNode": { + "id": -10, + "bounding": [ + 270, + 3100, + 120, + 240 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + 3620, + 3120, + 120, + 60 + ] + }, + "inputs": [ + { + "id": "6fe179c4-d96f-4383-b202-844f6de4922e", + "name": "input", + "type": "IMAGE,MASK", + "linkIds": [ + 251 + ], + "localized_name": "input", + "label": "first_frame", + "pos": [ + 370, + 3120 + ] + }, + { + "id": "e80df1ae-5f39-4f86-91bd-0467635e2f2d", + "name": "input_1", + "type": "IMAGE,MASK", + "linkIds": [ + 253 + ], + "localized_name": "input_1", + "label": "last_frame", + "pos": [ + 370, + 3140 + ] + }, + { + "id": "433148fa-bf73-4ab1-81d9-09e2e38ed861", + "name": "text", + "type": "STRING", + "linkIds": [ + 265 + ], + "pos": [ + 370, + 3160 + ] + }, + { + "id": "36915bc8-a6ed-4d48-8619-e0e8723228e9", + "name": "value", + "type": "INT", + "linkIds": [ + 266 + ], + "label": "width", + "pos": [ + 370, + 3180 + ] + }, + { + "id": "425a36b8-91ab-41b7-81e9-496eba064ec8", + "name": "value_1", + "type": "INT", + "linkIds": [ + 267 + ], + "label": "height", + "pos": [ + 370, + 3200 + ] + }, + { + "id": "0c9e003b-bd07-4b7d-aa6d-789e138ed161", + "name": "value_2", + "type": "INT", + "linkIds": [ + 268 + ], + "label": "duration", + "pos": [ + 370, + 3220 + ] + }, + { + "id": "581b52ff-21c5-4774-ac2a-8f69a7e09e2e", + "name": "value_3", + "type": "INT", + "linkIds": [ + 269 + ], + "label": "fps", + "pos": [ + 370, + 3240 + ] + }, + { + "id": "d03cc171-45da-4658-99aa-77252bbcf522", + "name": "noise_seed", + "type": "INT", + "linkIds": [ + 270 + ], + "pos": [ + 370, + 3260 + ] + }, + { + "id": "e68e61c8-905e-43ac-8c76-65ac52270a08", + "name": "ckpt_name_1", + "type": "COMBO", + "linkIds": [ + 272, + 275, + 276 + ], + "label": "ckpt_name", + "pos": [ + 370, + 3280 + ] + }, + { + "id": "5d065f3b-891b-499f-950b-c2df0be24536", + "name": "text_encoder", + "type": "COMBO", + "linkIds": [ + 273 + ], + "pos": [ + 370, + 3300 + ] + } + ], + "outputs": [ + { + "id": "0c8c2dc0-c67c-4bc2-9e57-6aa00db2e3a9", + "name": "VIDEO", + "type": "VIDEO", + "linkIds": [ + 252 + ], + "localized_name": "VIDEO", + "pos": [ + 3640, + 3140 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 195, + "type": "LTXVPreprocess", + "pos": [ + 1480, + 3780 + ], + "size": [ + 230, + 110 + ], + "flags": { + "collapsed": false + }, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 203 + }, + { + "localized_name": "img_compression", + "name": "img_compression", + "type": "INT", + "widget": { + "name": "img_compression" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "output_image", + "name": "output_image", + "type": "IMAGE", + "links": [ + 229 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "LTXVPreprocess", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 25 + ] + }, + { + "id": 196, + "type": "RandomNoise", + "pos": [ + 1990, + 2320 + ], + "size": [ + 280, + 110 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "noise_seed", + "name": "noise_seed", + "type": "INT", + "widget": { + "name": "noise_seed" + }, + "link": 270 + } + ], + "outputs": [ + { + "localized_name": "NOISE", + "name": "NOISE", + "type": "NOISE", + "links": [ + 246 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.14.1", + "ue_properties": { + "widget_ue_connectable": { + "noise_seed": true + }, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "RandomNoise", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 315253765879496, + "randomize" + ] + }, + { + "id": 197, + "type": "LTXVEmptyLatentAudio", + "pos": [ + 2090, + 3820 + ], + "size": [ + 280, + 170 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [ + { + "localized_name": "audio_vae", + "name": "audio_vae", + "type": "VAE", + "link": 205 + }, + { + "localized_name": "frames_number", + "name": "frames_number", + "type": "INT", + "widget": { + "name": "frames_number" + }, + "link": 262 + }, + { + "localized_name": "frame_rate", + "name": "frame_rate", + "type": "INT", + "widget": { + "name": "frame_rate" + }, + "link": 207 + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "widget": { + "name": "batch_size" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "Latent", + "name": "Latent", + "type": "LATENT", + "links": [ + 245 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.68", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "LTXVEmptyLatentAudio", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 97, + 25, + 1 + ] + }, + { + "id": 198, + "type": "PrimitiveInt", + "pos": [ + 760, + 3650 + ], + "size": [ + 230, + 110 + ], + "flags": {}, + "order": 5, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": 268 + } + ], + "outputs": [ + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 260 + ] + } + ], + "title": "Duration", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "PrimitiveInt", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 5, + "fixed" + ] + }, + { + "id": 199, + "type": "LTXVPreprocess", + "pos": [ + 1480, + 3340 + ], + "size": [ + 230, + 110 + ], + "flags": { + "collapsed": false + }, + "order": 6, + "mode": 0, + "inputs": [ + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 210 + }, + { + "localized_name": "img_compression", + "name": "img_compression", + "type": "INT", + "widget": { + "name": "img_compression" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "output_image", + "name": "output_image", + "type": "IMAGE", + "links": [ + 240 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "LTXVPreprocess", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 25 + ] + }, + { + "id": 200, + "type": "LTXVCropGuides", + "pos": [ + 2820, + 2450 + ], + "size": [ + 280, + 120 + ], + "flags": {}, + "order": 7, + "mode": 0, + "inputs": [ + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "link": 213 + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "link": 214 + }, + { + "localized_name": "latent", + "name": "latent", + "type": "LATENT", + "link": 215 + } + ], + "outputs": [ + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "links": [] + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "links": [] + }, + { + "localized_name": "latent", + "name": "latent", + "type": "LATENT", + "links": [ + 211 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.8.2", + "ue_properties": { + "widget_ue_connectable": {}, + "input_ue_unconnectable": {}, + "version": "7.5.2" + }, + "Node name for S&R": "LTXVCropGuides", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 201, + "type": "EmptyLTXVLatentVideo", + "pos": [ + 2090, + 3580 + ], + "size": [ + 280, + 200 + ], + "flags": {}, + "order": 8, + "mode": 0, + "inputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": 218 + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": 219 + }, + { + "localized_name": "length", + "name": "length", + "type": "INT", + "widget": { + "name": "length" + }, + "link": 263 + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "widget": { + "name": "batch_size" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "links": [ + 239 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.60", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "EmptyLTXVLatentVideo", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 768, + 512, + 97, + 1 + ] + }, + { + "id": 202, + "type": "LTXVConditioning", + "pos": [ + 2090, + 3400 + ], + "size": [ + 280, + 130 + ], + "flags": {}, + "order": 9, + "mode": 0, + "inputs": [ + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "link": 221 + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "link": 222 + }, + { + "localized_name": "frame_rate", + "name": "frame_rate", + "type": "FLOAT", + "widget": { + "name": "frame_rate" + }, + "link": 223 + } + ], + "outputs": [ + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "links": [ + 236 + ] + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "links": [ + 237 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.56", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "LTXVConditioning", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 25 + ] + }, + { + "id": 203, + "type": "GetImageSize", + "pos": [ + 1480, + 3500 + ], + "size": [ + 230, + 130 + ], + "flags": {}, + "order": 10, + "mode": 0, + "inputs": [ + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 224 + } + ], + "outputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "links": [ + 218 + ] + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "links": [ + 219 + ] + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "links": [] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.14.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "GetImageSize", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 204, + "type": "LTXVAddGuide", + "pos": [ + 2750, + 3700 + ], + "size": [ + 280, + 240 + ], + "flags": {}, + "order": 11, + "mode": 0, + "inputs": [ + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "link": 225 + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "link": 226 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 227 + }, + { + "localized_name": "latent", + "name": "latent", + "type": "LATENT", + "link": 228 + }, + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 229 + }, + { + "localized_name": "frame_idx", + "name": "frame_idx", + "type": "INT", + "widget": { + "name": "frame_idx" + }, + "link": null + }, + { + "localized_name": "strength", + "name": "strength", + "type": "FLOAT", + "widget": { + "name": "strength" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "links": [ + 213, + 242 + ] + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "links": [ + 214, + 243 + ] + }, + { + "localized_name": "latent", + "name": "latent", + "type": "LATENT", + "links": [ + 244 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.12.3", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "LTXVAddGuide", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + -1, + 0.7 + ] + }, + { + "id": 205, + "type": "PrimitiveInt", + "pos": [ + 760, + 3800 + ], + "size": [ + 230, + 110 + ], + "flags": {}, + "order": 12, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": 269 + } + ], + "outputs": [ + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 207, + 235, + 261 + ] + } + ], + "title": "Frame Rate(int)", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "PrimitiveInt", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 25, + "fixed" + ] + }, + { + "id": 206, + "type": "LTXVAddGuide", + "pos": [ + 2750, + 3430 + ], + "size": [ + 280, + 240 + ], + "flags": {}, + "order": 13, + "mode": 0, + "inputs": [ + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "link": 236 + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "link": 237 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 238 + }, + { + "localized_name": "latent", + "name": "latent", + "type": "LATENT", + "link": 239 + }, + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 240 + }, + { + "localized_name": "frame_idx", + "name": "frame_idx", + "type": "INT", + "widget": { + "name": "frame_idx" + }, + "link": null + }, + { + "localized_name": "strength", + "name": "strength", + "type": "FLOAT", + "widget": { + "name": "strength" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "links": [ + 225 + ] + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "links": [ + 226 + ] + }, + { + "localized_name": "latent", + "name": "latent", + "type": "LATENT", + "links": [ + 228 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.12.3", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "LTXVAddGuide", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 0, + 0.7 + ] + }, + { + "id": 207, + "type": "CFGGuider", + "pos": [ + 1990, + 2500 + ], + "size": [ + 280, + 160 + ], + "flags": {}, + "order": 14, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 241 + }, + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "link": 242 + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "link": 243 + }, + { + "localized_name": "cfg", + "name": "cfg", + "type": "FLOAT", + "widget": { + "name": "cfg" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "GUIDER", + "name": "GUIDER", + "type": "GUIDER", + "links": [ + 247 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.14.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "CFGGuider", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 1 + ] + }, + { + "id": 208, + "type": "SamplerEulerAncestral", + "pos": [ + 1990, + 2720 + ], + "size": [ + 280, + 120 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "localized_name": "eta", + "name": "eta", + "type": "FLOAT", + "widget": { + "name": "eta" + }, + "link": null + }, + { + "localized_name": "s_noise", + "name": "s_noise", + "type": "FLOAT", + "widget": { + "name": "s_noise" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "SAMPLER", + "name": "SAMPLER", + "type": "SAMPLER", + "links": [ + 248 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.14.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "SamplerEulerAncestral", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 0, + 1 + ] + }, + { + "id": 209, + "type": "ManualSigmas", + "pos": [ + 1990, + 2910 + ], + "size": [ + 280, + 110 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "sigmas", + "name": "sigmas", + "type": "STRING", + "widget": { + "name": "sigmas" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "SIGMAS", + "name": "SIGMAS", + "type": "SIGMAS", + "links": [ + 249 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.14.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ManualSigmas", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "1., 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0" + ] + }, + { + "id": 210, + "type": "LTXVConcatAVLatent", + "pos": [ + 1990, + 3090 + ], + "size": [ + 280, + 100 + ], + "flags": {}, + "order": 15, + "mode": 0, + "inputs": [ + { + "localized_name": "video_latent", + "name": "video_latent", + "type": "LATENT", + "link": 244 + }, + { + "localized_name": "audio_latent", + "name": "audio_latent", + "type": "LATENT", + "link": 245 + } + ], + "outputs": [ + { + "localized_name": "latent", + "name": "latent", + "type": "LATENT", + "links": [ + 250 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "LTXVConcatAVLatent", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 211, + "type": "SamplerCustomAdvanced", + "pos": [ + 2460, + 2330 + ], + "size": [ + 230, + 170 + ], + "flags": {}, + "order": 16, + "mode": 0, + "inputs": [ + { + "localized_name": "noise", + "name": "noise", + "type": "NOISE", + "link": 246 + }, + { + "localized_name": "guider", + "name": "guider", + "type": "GUIDER", + "link": 247 + }, + { + "localized_name": "sampler", + "name": "sampler", + "type": "SAMPLER", + "link": 248 + }, + { + "localized_name": "sigmas", + "name": "sigmas", + "type": "SIGMAS", + "link": 249 + }, + { + "localized_name": "latent_image", + "name": "latent_image", + "type": "LATENT", + "link": 250 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "LATENT", + "links": [] + }, + { + "localized_name": "denoised_output", + "name": "denoised_output", + "type": "LATENT", + "links": [ + 204 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.14.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "SamplerCustomAdvanced", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 212, + "type": "ComfyMathExpression", + "pos": [ + 760, + 3970 + ], + "size": [ + 230, + 170 + ], + "flags": { + "collapsed": true + }, + "order": 17, + "mode": 0, + "inputs": [ + { + "label": "a", + "localized_name": "values.a", + "name": "values.a", + "type": "FLOAT,INT", + "link": 235 + }, + { + "label": "b", + "localized_name": "values.b", + "name": "values.b", + "shape": 7, + "type": "FLOAT,INT", + "link": null + }, + { + "localized_name": "expression", + "name": "expression", + "type": "STRING", + "widget": { + "name": "expression" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "FLOAT", + "name": "FLOAT", + "type": "FLOAT", + "links": [ + 223, + 234 + ] + }, + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.17.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ComfyMathExpression", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "a" + ] + }, + { + "id": 213, + "type": "ResizeImageMaskNode", + "pos": [ + 1130, + 3340 + ], + "size": [ + 280, + 160 + ], + "flags": {}, + "order": 18, + "mode": 0, + "inputs": [ + { + "localized_name": "input", + "name": "input", + "type": "IMAGE,MASK", + "link": 251 + }, + { + "localized_name": "resize_type", + "name": "resize_type", + "type": "COMFY_DYNAMICCOMBO_V3", + "widget": { + "name": "resize_type" + }, + "link": null + }, + { + "localized_name": "width", + "name": "resize_type.width", + "type": "INT", + "widget": { + "name": "resize_type.width" + }, + "link": 208 + }, + { + "localized_name": "height", + "name": "resize_type.height", + "type": "INT", + "widget": { + "name": "resize_type.height" + }, + "link": 209 + }, + { + "localized_name": "crop", + "name": "resize_type.crop", + "type": "COMBO", + "widget": { + "name": "resize_type.crop" + }, + "link": null + }, + { + "localized_name": "scale_method", + "name": "scale_method", + "type": "COMBO", + "widget": { + "name": "scale_method" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "resized", + "name": "resized", + "type": "*", + "links": [ + 210, + 224 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.14.1", + "ue_properties": { + "widget_ue_connectable": { + "resize_type.width": true, + "resize_type.height": true + }, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ResizeImageMaskNode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "scale dimensions", + 640, + 360, + "center", + "nearest-exact" + ] + }, + { + "id": 214, + "type": "ResizeImageMaskNode", + "pos": [ + 1130, + 3780 + ], + "size": [ + 280, + 160 + ], + "flags": {}, + "order": 19, + "mode": 0, + "inputs": [ + { + "localized_name": "input", + "name": "input", + "type": "IMAGE,MASK", + "link": 253 + }, + { + "localized_name": "resize_type", + "name": "resize_type", + "type": "COMFY_DYNAMICCOMBO_V3", + "widget": { + "name": "resize_type" + }, + "link": null + }, + { + "localized_name": "width", + "name": "resize_type.width", + "type": "INT", + "widget": { + "name": "resize_type.width" + }, + "link": 201 + }, + { + "localized_name": "height", + "name": "resize_type.height", + "type": "INT", + "widget": { + "name": "resize_type.height" + }, + "link": 202 + }, + { + "localized_name": "crop", + "name": "resize_type.crop", + "type": "COMBO", + "widget": { + "name": "resize_type.crop" + }, + "link": null + }, + { + "localized_name": "scale_method", + "name": "scale_method", + "type": "COMBO", + "widget": { + "name": "scale_method" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "resized", + "name": "resized", + "type": "*", + "links": [ + 203 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.14.1", + "ue_properties": { + "widget_ue_connectable": { + "resize_type.width": true, + "resize_type.height": true + }, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ResizeImageMaskNode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "scale dimensions", + 640, + 360, + "center", + "nearest-exact" + ] + }, + { + "id": 215, + "type": "PrimitiveInt", + "pos": [ + 760, + 3340 + ], + "size": [ + 230, + 110 + ], + "flags": {}, + "order": 20, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": 266 + } + ], + "outputs": [ + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 201, + 208 + ] + } + ], + "title": "Width", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "PrimitiveInt", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 1280, + "fixed" + ] + }, + { + "id": 216, + "type": "PrimitiveInt", + "pos": [ + 760, + 3490 + ], + "size": [ + 230, + 110 + ], + "flags": {}, + "order": 21, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": 267 + } + ], + "outputs": [ + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 202, + 209 + ] + } + ], + "title": "height", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "PrimitiveInt", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 720, + "fixed" + ] + }, + { + "id": 217, + "type": "CLIPTextEncode", + "pos": [ + 1320, + 2870 + ], + "size": [ + 590, + 200 + ], + "flags": { + "collapsed": false + }, + "order": 22, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 230 + }, + { + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 222 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.56", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "CLIPTextEncode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "blurry, out of focus, overexposed, underexposed, low contrast, washed out colors, excessive noise, grainy texture, poor lighting, flickering, motion blur, distorted proportions, unnatural skin tones, deformed facial features, asymmetrical face, missing facial features, extra limbs, disfigured hands, wrong hand count, artifacts around text, unreadable text on shirt or hat, incorrect lettering on cap (“PNTR”), incorrect t-shirt slogan (“JUST DO IT”), missing microphone, misplaced microphone, inconsistent perspective, camera shake, incorrect depth of field, background too sharp, background clutter, distracting reflections, harsh shadows, inconsistent lighting direction, color banding, cartoonish rendering, 3D CGI look, unrealistic materials, uncanny valley effect, incorrect ethnicity, wrong gender, exaggerated expressions, smiling, laughing, exaggerated sadness, wrong gaze direction, eyes looking at camera, mismatched lip sync, silent or muted audio, distorted voice, robotic voice, echo, background noise, off-sync audio, missing sniff sounds, incorrect dialogue, added dialogue, repetitive speech, jittery movement, awkward pauses, incorrect timing, unnatural transitions, inconsistent framing, tilted camera, missing door or shelves, missing shallow depth of field, flat lighting, inconsistent tone, cinematic oversaturation, stylized filters, or AI artifacts." + ], + "color": "#323", + "bgcolor": "#535" + }, + { + "id": 218, + "type": "CreateVideo", + "pos": [ + 3280, + 2320 + ], + "size": [ + 280, + 130 + ], + "flags": {}, + "order": 23, + "mode": 0, + "inputs": [ + { + "localized_name": "images", + "name": "images", + "type": "IMAGE", + "link": 232 + }, + { + "localized_name": "audio", + "name": "audio", + "shape": 7, + "type": "AUDIO", + "link": 233 + }, + { + "localized_name": "fps", + "name": "fps", + "type": "FLOAT", + "widget": { + "name": "fps" + }, + "link": 234 + } + ], + "outputs": [ + { + "localized_name": "VIDEO", + "name": "VIDEO", + "type": "VIDEO", + "links": [ + 252 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.14.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "CreateVideo", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 24 + ] + }, + { + "id": 219, + "type": "VAEDecodeTiled", + "pos": [ + 2820, + 2630 + ], + "size": [ + 280, + 200 + ], + "flags": {}, + "order": 24, + "mode": 0, + "inputs": [ + { + "localized_name": "samples", + "name": "samples", + "type": "LATENT", + "link": 211 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 212 + }, + { + "localized_name": "tile_size", + "name": "tile_size", + "type": "INT", + "widget": { + "name": "tile_size" + }, + "link": null + }, + { + "localized_name": "overlap", + "name": "overlap", + "type": "INT", + "widget": { + "name": "overlap" + }, + "link": null + }, + { + "localized_name": "temporal_size", + "name": "temporal_size", + "type": "INT", + "widget": { + "name": "temporal_size" + }, + "link": null + }, + { + "localized_name": "temporal_overlap", + "name": "temporal_overlap", + "type": "INT", + "widget": { + "name": "temporal_overlap" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [ + 232 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "VAEDecodeTiled", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 768, + 64, + 4096, + 64 + ] + }, + { + "id": 220, + "type": "LTXVAudioVAEDecode", + "pos": [ + 2820, + 2920 + ], + "size": [ + 280, + 100 + ], + "flags": {}, + "order": 25, + "mode": 0, + "inputs": [ + { + "localized_name": "samples", + "name": "samples", + "type": "LATENT", + "link": 216 + }, + { + "label": "Audio VAE", + "localized_name": "audio_vae", + "name": "audio_vae", + "type": "VAE", + "link": 217 + } + ], + "outputs": [ + { + "localized_name": "Audio", + "name": "Audio", + "type": "AUDIO", + "links": [ + 233 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "LTXVAudioVAEDecode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 221, + "type": "LTXVSeparateAVLatent", + "pos": [ + 2460, + 2580 + ], + "size": [ + 250, + 100 + ], + "flags": {}, + "order": 26, + "mode": 0, + "inputs": [ + { + "localized_name": "av_latent", + "name": "av_latent", + "type": "LATENT", + "link": 204 + } + ], + "outputs": [ + { + "localized_name": "video_latent", + "name": "video_latent", + "type": "LATENT", + "links": [ + 215 + ] + }, + { + "localized_name": "audio_latent", + "name": "audio_latent", + "type": "LATENT", + "links": [ + 216 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.5.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "LTXVSeparateAVLatent", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 222, + "type": "CLIPTextEncode", + "pos": [ + 1310, + 2380 + ], + "size": [ + 620, + 420 + ], + "flags": {}, + "order": 27, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 231 + }, + { + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": 265 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 221 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.56", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.5.2", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "CLIPTextEncode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "" + ], + "color": "#232", + "bgcolor": "#353" + }, + { + "id": 223, + "type": "CheckpointLoaderSimple", + "pos": [ + 770, + 2380 + ], + "size": [ + 420, + 160 + ], + "flags": {}, + "order": 28, + "mode": 0, + "inputs": [ + { + "localized_name": "ckpt_name", + "name": "ckpt_name", + "type": "COMBO", + "widget": { + "name": "ckpt_name" + }, + "link": 276 + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 241 + ] + }, + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [] + }, + { + "localized_name": "VAE", + "name": "VAE", + "type": "VAE", + "links": [ + 212, + 227, + 238 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.10.0", + "ue_properties": { + "widget_ue_connectable": {}, + "input_ue_unconnectable": {}, + "version": "7.5.2" + }, + "Node name for S&R": "CheckpointLoaderSimple", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "ltx-2.3-22b-distilled-fp8.safetensors", + "url": "https://huggingface.co/Lightricks/LTX-2.3-fp8/resolve/main/ltx-2.3-22b-distilled-fp8.safetensors", + "directory": "checkpoints" + } + ] + }, + "widgets_values": [ + "ltx-2.3-22b-distilled-fp8.safetensors" + ] + }, + { + "id": 224, + "type": "LTXVAudioVAELoader", + "pos": [ + 770, + 2660 + ], + "size": [ + 420, + 110 + ], + "flags": {}, + "order": 29, + "mode": 0, + "inputs": [ + { + "localized_name": "ckpt_name", + "name": "ckpt_name", + "type": "COMBO", + "widget": { + "name": "ckpt_name" + }, + "link": 272 + } + ], + "outputs": [ + { + "localized_name": "Audio VAE", + "name": "Audio VAE", + "type": "VAE", + "links": [ + 205, + 217 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.10.0", + "ue_properties": { + "widget_ue_connectable": {}, + "input_ue_unconnectable": {}, + "version": "7.5.2" + }, + "Node name for S&R": "LTXVAudioVAELoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "ltx-2.3-22b-distilled-fp8.safetensors", + "url": "https://huggingface.co/Lightricks/LTX-2.3-fp8/resolve/main/ltx-2.3-22b-distilled-fp8.safetensors", + "directory": "checkpoints" + } + ] + }, + "widgets_values": [ + "ltx-2.3-22b-distilled-fp8.safetensors" + ] + }, + { + "id": 225, + "type": "LTXAVTextEncoderLoader", + "pos": [ + 770, + 2890 + ], + "size": [ + 410, + 160 + ], + "flags": {}, + "order": 30, + "mode": 0, + "inputs": [ + { + "localized_name": "text_encoder", + "name": "text_encoder", + "type": "COMBO", + "widget": { + "name": "text_encoder" + }, + "link": 273 + }, + { + "localized_name": "ckpt_name", + "name": "ckpt_name", + "type": "COMBO", + "widget": { + "name": "ckpt_name" + }, + "link": 275 + }, + { + "localized_name": "device", + "name": "device", + "type": "COMBO", + "widget": { + "name": "device" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [ + 230, + 231 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.10.0", + "ue_properties": { + "widget_ue_connectable": {}, + "input_ue_unconnectable": {}, + "version": "7.5.2" + }, + "Node name for S&R": "LTXAVTextEncoderLoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "gemma_3_12B_it_fp4_mixed.safetensors", + "url": "https://huggingface.co/Comfy-Org/ltx-2/resolve/main/split_files/text_encoders/gemma_3_12B_it_fp4_mixed.safetensors", + "directory": "text_encoders" + }, + { + "name": "ltx-2.3-22b-distilled-fp8.safetensors", + "url": "https://huggingface.co/Lightricks/LTX-2.3-fp8/resolve/main/ltx-2.3-22b-distilled-fp8.safetensors", + "directory": "checkpoints" + } + ] + }, + "widgets_values": [ + "gemma_3_12B_it_fp4_mixed.safetensors", + "ltx-2.3-22b-distilled-fp8.safetensors", + "default" + ] + }, + { + "id": 226, + "type": "ComfyMathExpression", + "pos": [ + 760, + 4020 + ], + "size": [ + 400, + 200 + ], + "flags": { + "collapsed": true + }, + "order": 31, + "mode": 0, + "inputs": [ + { + "label": "a", + "localized_name": "values.a", + "name": "values.a", + "type": "FLOAT,INT", + "link": 260 + }, + { + "label": "b", + "localized_name": "values.b", + "name": "values.b", + "shape": 7, + "type": "FLOAT,INT", + "link": 261 + }, + { + "label": "c", + "localized_name": "values.c", + "name": "values.c", + "shape": 7, + "type": "FLOAT,INT", + "link": null + }, + { + "localized_name": "expression", + "name": "expression", + "type": "STRING", + "widget": { + "name": "expression" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "FLOAT", + "name": "FLOAT", + "type": "FLOAT", + "links": null + }, + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 262, + 263 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.18.1", + "ue_properties": { + "widget_ue_connectable": {}, + "input_ue_unconnectable": {}, + "version": "7.7" + }, + "Node name for S&R": "ComfyMathExpression" + }, + "widgets_values": [ + "a * b + 1" + ] + } + ], + "groups": [ + { + "id": 1, + "title": "Conditioning", + "bounding": [ + 1850, + 3250, + 1370, + 800 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 2, + "title": "Settings", + "bounding": [ + 730, + 3250, + 290, + 800 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 3, + "title": "FIrst Frame", + "bounding": [ + 1050, + 3250, + 770, + 400 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 4, + "title": "Last Frame", + "bounding": [ + 1050, + 3680, + 770, + 370 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 5, + "title": "Model", + "bounding": [ + 730, + 2240, + 500, + 980 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 6, + "title": "Prompt", + "bounding": [ + 1260, + 2240, + 680, + 980 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 7, + "title": "Sampling", + "bounding": [ + 1970, + 2240, + 770, + 980 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 8, + "title": "Decoding", + "bounding": [ + 2770, + 2240, + 450, + 980 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + } + ], + "links": [ + { + "id": 203, + "origin_id": 214, + "origin_slot": 0, + "target_id": 195, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 205, + "origin_id": 224, + "origin_slot": 0, + "target_id": 197, + "target_slot": 0, + "type": "VAE" + }, + { + "id": 207, + "origin_id": 205, + "origin_slot": 0, + "target_id": 197, + "target_slot": 2, + "type": "INT" + }, + { + "id": 210, + "origin_id": 213, + "origin_slot": 0, + "target_id": 199, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 213, + "origin_id": 204, + "origin_slot": 0, + "target_id": 200, + "target_slot": 0, + "type": "CONDITIONING" + }, + { + "id": 214, + "origin_id": 204, + "origin_slot": 1, + "target_id": 200, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 215, + "origin_id": 221, + "origin_slot": 0, + "target_id": 200, + "target_slot": 2, + "type": "LATENT" + }, + { + "id": 218, + "origin_id": 203, + "origin_slot": 0, + "target_id": 201, + "target_slot": 0, + "type": "INT" + }, + { + "id": 219, + "origin_id": 203, + "origin_slot": 1, + "target_id": 201, + "target_slot": 1, + "type": "INT" + }, + { + "id": 221, + "origin_id": 222, + "origin_slot": 0, + "target_id": 202, + "target_slot": 0, + "type": "CONDITIONING" + }, + { + "id": 222, + "origin_id": 217, + "origin_slot": 0, + "target_id": 202, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 223, + "origin_id": 212, + "origin_slot": 0, + "target_id": 202, + "target_slot": 2, + "type": "FLOAT" + }, + { + "id": 224, + "origin_id": 213, + "origin_slot": 0, + "target_id": 203, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 225, + "origin_id": 206, + "origin_slot": 0, + "target_id": 204, + "target_slot": 0, + "type": "CONDITIONING" + }, + { + "id": 226, + "origin_id": 206, + "origin_slot": 1, + "target_id": 204, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 227, + "origin_id": 223, + "origin_slot": 2, + "target_id": 204, + "target_slot": 2, + "type": "VAE" + }, + { + "id": 228, + "origin_id": 206, + "origin_slot": 2, + "target_id": 204, + "target_slot": 3, + "type": "LATENT" + }, + { + "id": 229, + "origin_id": 195, + "origin_slot": 0, + "target_id": 204, + "target_slot": 4, + "type": "IMAGE" + }, + { + "id": 236, + "origin_id": 202, + "origin_slot": 0, + "target_id": 206, + "target_slot": 0, + "type": "CONDITIONING" + }, + { + "id": 237, + "origin_id": 202, + "origin_slot": 1, + "target_id": 206, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 238, + "origin_id": 223, + "origin_slot": 2, + "target_id": 206, + "target_slot": 2, + "type": "VAE" + }, + { + "id": 239, + "origin_id": 201, + "origin_slot": 0, + "target_id": 206, + "target_slot": 3, + "type": "LATENT" + }, + { + "id": 240, + "origin_id": 199, + "origin_slot": 0, + "target_id": 206, + "target_slot": 4, + "type": "IMAGE" + }, + { + "id": 241, + "origin_id": 223, + "origin_slot": 0, + "target_id": 207, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 242, + "origin_id": 204, + "origin_slot": 0, + "target_id": 207, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 243, + "origin_id": 204, + "origin_slot": 1, + "target_id": 207, + "target_slot": 2, + "type": "CONDITIONING" + }, + { + "id": 244, + "origin_id": 204, + "origin_slot": 2, + "target_id": 210, + "target_slot": 0, + "type": "LATENT" + }, + { + "id": 245, + "origin_id": 197, + "origin_slot": 0, + "target_id": 210, + "target_slot": 1, + "type": "LATENT" + }, + { + "id": 246, + "origin_id": 196, + "origin_slot": 0, + "target_id": 211, + "target_slot": 0, + "type": "NOISE" + }, + { + "id": 247, + "origin_id": 207, + "origin_slot": 0, + "target_id": 211, + "target_slot": 1, + "type": "GUIDER" + }, + { + "id": 248, + "origin_id": 208, + "origin_slot": 0, + "target_id": 211, + "target_slot": 2, + "type": "SAMPLER" + }, + { + "id": 249, + "origin_id": 209, + "origin_slot": 0, + "target_id": 211, + "target_slot": 3, + "type": "SIGMAS" + }, + { + "id": 250, + "origin_id": 210, + "origin_slot": 0, + "target_id": 211, + "target_slot": 4, + "type": "LATENT" + }, + { + "id": 235, + "origin_id": 205, + "origin_slot": 0, + "target_id": 212, + "target_slot": 0, + "type": "INT" + }, + { + "id": 208, + "origin_id": 215, + "origin_slot": 0, + "target_id": 213, + "target_slot": 2, + "type": "INT" + }, + { + "id": 209, + "origin_id": 216, + "origin_slot": 0, + "target_id": 213, + "target_slot": 3, + "type": "INT" + }, + { + "id": 201, + "origin_id": 215, + "origin_slot": 0, + "target_id": 214, + "target_slot": 2, + "type": "INT" + }, + { + "id": 202, + "origin_id": 216, + "origin_slot": 0, + "target_id": 214, + "target_slot": 3, + "type": "INT" + }, + { + "id": 230, + "origin_id": 225, + "origin_slot": 0, + "target_id": 217, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 232, + "origin_id": 219, + "origin_slot": 0, + "target_id": 218, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 233, + "origin_id": 220, + "origin_slot": 0, + "target_id": 218, + "target_slot": 1, + "type": "AUDIO" + }, + { + "id": 234, + "origin_id": 212, + "origin_slot": 0, + "target_id": 218, + "target_slot": 2, + "type": "FLOAT" + }, + { + "id": 211, + "origin_id": 200, + "origin_slot": 2, + "target_id": 219, + "target_slot": 0, + "type": "LATENT" + }, + { + "id": 212, + "origin_id": 223, + "origin_slot": 2, + "target_id": 219, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 216, + "origin_id": 221, + "origin_slot": 1, + "target_id": 220, + "target_slot": 0, + "type": "LATENT" + }, + { + "id": 217, + "origin_id": 224, + "origin_slot": 0, + "target_id": 220, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 204, + "origin_id": 211, + "origin_slot": 1, + "target_id": 221, + "target_slot": 0, + "type": "LATENT" + }, + { + "id": 231, + "origin_id": 225, + "origin_slot": 0, + "target_id": 222, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 251, + "origin_id": -10, + "origin_slot": 0, + "target_id": 213, + "target_slot": 0, + "type": "IMAGE,MASK" + }, + { + "id": 253, + "origin_id": -10, + "origin_slot": 1, + "target_id": 214, + "target_slot": 0, + "type": "IMAGE,MASK" + }, + { + "id": 252, + "origin_id": 218, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "VIDEO" + }, + { + "id": 260, + "origin_id": 198, + "origin_slot": 0, + "target_id": 226, + "target_slot": 0, + "type": "INT" + }, + { + "id": 261, + "origin_id": 205, + "origin_slot": 0, + "target_id": 226, + "target_slot": 1, + "type": "INT" + }, + { + "id": 262, + "origin_id": 226, + "origin_slot": 1, + "target_id": 197, + "target_slot": 1, + "type": "INT" + }, + { + "id": 263, + "origin_id": 226, + "origin_slot": 1, + "target_id": 201, + "target_slot": 2, + "type": "INT" + }, + { + "id": 265, + "origin_id": -10, + "origin_slot": 2, + "target_id": 222, + "target_slot": 1, + "type": "STRING" + }, + { + "id": 266, + "origin_id": -10, + "origin_slot": 3, + "target_id": 215, + "target_slot": 0, + "type": "INT" + }, + { + "id": 267, + "origin_id": -10, + "origin_slot": 4, + "target_id": 216, + "target_slot": 0, + "type": "INT" + }, + { + "id": 268, + "origin_id": -10, + "origin_slot": 5, + "target_id": 198, + "target_slot": 0, + "type": "INT" + }, + { + "id": 269, + "origin_id": -10, + "origin_slot": 6, + "target_id": 205, + "target_slot": 0, + "type": "INT" + }, + { + "id": 270, + "origin_id": -10, + "origin_slot": 7, + "target_id": 196, + "target_slot": 0, + "type": "INT" + }, + { + "id": 272, + "origin_id": -10, + "origin_slot": 8, + "target_id": 224, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 273, + "origin_id": -10, + "origin_slot": 9, + "target_id": 225, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 275, + "origin_id": -10, + "origin_slot": 8, + "target_id": 225, + "target_slot": 1, + "type": "COMBO" + }, + { + "id": 276, + "origin_id": -10, + "origin_slot": 8, + "target_id": 223, + "target_slot": 0, + "type": "COMBO" + } + ], + "extra": {}, + "category": "Video generation and editing/First-Last-Frame to Video", + "description": "Generates a video that interpolates between the first and last keyframes using LTX-2.3, including optional audio." + } + ] + }, + "extra": { + "ue_links": [] + } +} \ No newline at end of file diff --git a/blueprints/Frame Interpolation.json b/blueprints/Frame Interpolation.json new file mode 100644 index 000000000..8e183de7e --- /dev/null +++ b/blueprints/Frame Interpolation.json @@ -0,0 +1,858 @@ +{ + "revision": 0, + "last_node_id": 16, + "last_link_id": 0, + "nodes": [ + { + "id": 16, + "type": "022693be-2baa-4009-870a-28921508a7ef", + "pos": [ + -2990, + -3240 + ], + "size": [ + 410, + 200 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "video", + "name": "video", + "type": "VIDEO", + "link": null + }, + { + "label": "multiplier", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": null + }, + { + "label": "enable_fps_multiplier", + "name": "value_1", + "type": "BOOLEAN", + "widget": { + "name": "value_1" + }, + "link": null + }, + { + "name": "model_name", + "type": "COMBO", + "widget": { + "name": "model_name" + }, + "link": null + } + ], + "outputs": [ + { + "label": "VIDEO", + "name": "VIDEO_1", + "type": "VIDEO", + "links": [] + }, + { + "name": "IMAGE", + "type": "IMAGE", + "links": null + } + ], + "properties": { + "proxyWidgets": [ + [ + "9", + "value" + ], + [ + "13", + "value" + ], + [ + "1", + "model_name" + ] + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "cnr_id": "comfy-core", + "ver": "0.19.3" + }, + "widgets_values": [], + "title": "Frame Interpolation" + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "022693be-2baa-4009-870a-28921508a7ef", + "version": 1, + "state": { + "lastGroupId": 0, + "lastNodeId": 17, + "lastLinkId": 28, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "Frame Interpolation", + "inputNode": { + "id": -10, + "bounding": [ + -2810, + -3070, + 159.7421875, + 120 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + -1270, + -3075, + 120, + 80 + ] + }, + "inputs": [ + { + "id": "05e31c51-dcb6-4a1e-9651-1b9ad4f7a287", + "name": "video", + "type": "VIDEO", + "linkIds": [ + 2 + ], + "localized_name": "video", + "pos": [ + -2670.2578125, + -3050 + ] + }, + { + "id": "feecb409-7d1c-4a99-9c63-50c5fecdd3c9", + "name": "value", + "type": "INT", + "linkIds": [ + 22 + ], + "label": "multiplier", + "pos": [ + -2670.2578125, + -3030 + ] + }, + { + "id": "0b8a861b-b581-4068-9e8c-f8d15daf1ca6", + "name": "value_1", + "type": "BOOLEAN", + "linkIds": [ + 23 + ], + "label": "enable_fps_multiplier", + "pos": [ + -2670.2578125, + -3010 + ] + }, + { + "id": "a22b101e-8773-4e17-a297-7ee3aae09162", + "name": "model_name", + "type": "COMBO", + "linkIds": [ + 24 + ], + "pos": [ + -2670.2578125, + -2990 + ] + } + ], + "outputs": [ + { + "id": "ef2ada05-d5aa-492a-9394-6c3e71e39ebb", + "name": "VIDEO_1", + "type": "VIDEO", + "linkIds": [ + 26 + ], + "label": "VIDEO", + "pos": [ + -1250, + -3055 + ] + }, + { + "id": "5aacc622-2a07-4983-b31c-e04461f7f953", + "name": "IMAGE", + "type": "IMAGE", + "linkIds": [ + 28 + ], + "pos": [ + -1250, + -3035 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 1, + "type": "FrameInterpolationModelLoader", + "pos": [ + -2510, + -3370 + ], + "size": [ + 370, + 90 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "localized_name": "model_name", + "name": "model_name", + "type": "COMBO", + "widget": { + "name": "model_name" + }, + "link": 24 + } + ], + "outputs": [ + { + "localized_name": "INTERP_MODEL", + "name": "INTERP_MODEL", + "type": "INTERP_MODEL", + "links": [ + 1 + ] + } + ], + "properties": { + "Node name for S&R": "FrameInterpolationModelLoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "cnr_id": "comfy-core", + "ver": "0.19.3", + "models": [ + { + "name": "film_net_fp16.safetensors", + "url": "https://huggingface.co/Comfy-Org/frame_interpolation/resolve/main/frame_interpolation/film_net_fp16.safetensors", + "directory": "frame_interpolation" + } + ] + }, + "widgets_values": [ + "film_net_fp16.safetensors" + ] + }, + { + "id": 2, + "type": "FrameInterpolate", + "pos": [ + -2040, + -3370 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "interp_model", + "name": "interp_model", + "type": "INTERP_MODEL", + "link": 1 + }, + { + "localized_name": "images", + "name": "images", + "type": "IMAGE", + "link": 3 + }, + { + "localized_name": "multiplier", + "name": "multiplier", + "type": "INT", + "widget": { + "name": "multiplier" + }, + "link": 8 + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [ + 4, + 28 + ] + } + ], + "properties": { + "Node name for S&R": "FrameInterpolate", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "cnr_id": "comfy-core", + "ver": "0.19.3" + }, + "widgets_values": [ + 2 + ] + }, + { + "id": 5, + "type": "CreateVideo", + "pos": [ + -1600, + -3370 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "images", + "name": "images", + "type": "IMAGE", + "link": 4 + }, + { + "localized_name": "audio", + "name": "audio", + "shape": 7, + "type": "AUDIO", + "link": 5 + }, + { + "localized_name": "fps", + "name": "fps", + "type": "FLOAT", + "widget": { + "name": "fps" + }, + "link": 12 + } + ], + "outputs": [ + { + "localized_name": "VIDEO", + "name": "VIDEO", + "type": "VIDEO", + "links": [ + 26 + ] + } + ], + "properties": { + "Node name for S&R": "CreateVideo", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "cnr_id": "comfy-core", + "ver": "0.19.3" + }, + "widgets_values": [ + 30 + ] + }, + { + "id": 9, + "type": "PrimitiveInt", + "pos": [ + -2500, + -2970 + ], + "size": [ + 270, + 90 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": 22 + } + ], + "outputs": [ + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 8, + 19 + ] + } + ], + "title": "Int (Multiplier)", + "properties": { + "Node name for S&R": "PrimitiveInt", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "cnr_id": "comfy-core", + "ver": "0.19.3" + }, + "widgets_values": [ + 2, + "fixed" + ] + }, + { + "id": 10, + "type": "ComfySwitchNode", + "pos": [ + -1610, + -3120 + ], + "size": [ + 270, + 130 + ], + "flags": {}, + "order": 5, + "mode": 0, + "inputs": [ + { + "localized_name": "on_false", + "name": "on_false", + "type": "*", + "link": 11 + }, + { + "localized_name": "on_true", + "name": "on_true", + "type": "*", + "link": 13 + }, + { + "localized_name": "switch", + "name": "switch", + "type": "BOOLEAN", + "widget": { + "name": "switch" + }, + "link": 15 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "*", + "links": [ + 12 + ] + } + ], + "properties": { + "Node name for S&R": "ComfySwitchNode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "cnr_id": "comfy-core", + "ver": "0.19.3" + }, + "widgets_values": [ + true + ] + }, + { + "id": 13, + "type": "PrimitiveBoolean", + "pos": [ + -2500, + -2770 + ], + "size": [ + 310, + 90 + ], + "flags": {}, + "order": 7, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "BOOLEAN", + "widget": { + "name": "value" + }, + "link": 23 + } + ], + "outputs": [ + { + "localized_name": "BOOLEAN", + "name": "BOOLEAN", + "type": "BOOLEAN", + "links": [ + 15 + ] + } + ], + "title": "Boolean (Apply multiplier to FPS?)", + "properties": { + "Node name for S&R": "PrimitiveBoolean", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "cnr_id": "comfy-core", + "ver": "0.19.3" + }, + "widgets_values": [ + true + ] + }, + { + "id": 3, + "type": "GetVideoComponents", + "pos": [ + -2500, + -3170 + ], + "size": [ + 230, + 100 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "video", + "name": "video", + "type": "VIDEO", + "link": 2 + } + ], + "outputs": [ + { + "localized_name": "images", + "name": "images", + "type": "IMAGE", + "links": [ + 3 + ] + }, + { + "localized_name": "audio", + "name": "audio", + "type": "AUDIO", + "links": [ + 5 + ] + }, + { + "localized_name": "fps", + "name": "fps", + "type": "FLOAT", + "links": [ + 11, + 18 + ] + } + ], + "properties": { + "Node name for S&R": "GetVideoComponents", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "cnr_id": "comfy-core", + "ver": "0.19.3" + } + }, + { + "id": 11, + "type": "ComfyMathExpression", + "pos": [ + -2090, + -3070 + ], + "size": [ + 400, + 210 + ], + "flags": { + "collapsed": false + }, + "order": 6, + "mode": 0, + "inputs": [ + { + "label": "a", + "localized_name": "values.a", + "name": "values.a", + "type": "FLOAT,INT", + "link": 18 + }, + { + "label": "b", + "localized_name": "values.b", + "name": "values.b", + "shape": 7, + "type": "FLOAT,INT", + "link": 19 + }, + { + "label": "c", + "localized_name": "values.c", + "name": "values.c", + "shape": 7, + "type": "FLOAT,INT", + "link": null + }, + { + "localized_name": "expression", + "name": "expression", + "type": "STRING", + "widget": { + "name": "expression" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "FLOAT", + "name": "FLOAT", + "type": "FLOAT", + "links": [ + 13 + ] + }, + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": null + } + ], + "properties": { + "Node name for S&R": "ComfyMathExpression", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "cnr_id": "comfy-core", + "ver": "0.19.3" + }, + "widgets_values": [ + "min(abs(b), 16) * a" + ] + } + ], + "groups": [], + "links": [ + { + "id": 1, + "origin_id": 1, + "origin_slot": 0, + "target_id": 2, + "target_slot": 0, + "type": "INTERP_MODEL" + }, + { + "id": 3, + "origin_id": 3, + "origin_slot": 0, + "target_id": 2, + "target_slot": 1, + "type": "IMAGE" + }, + { + "id": 8, + "origin_id": 9, + "origin_slot": 0, + "target_id": 2, + "target_slot": 2, + "type": "INT" + }, + { + "id": 4, + "origin_id": 2, + "origin_slot": 0, + "target_id": 5, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 5, + "origin_id": 3, + "origin_slot": 1, + "target_id": 5, + "target_slot": 1, + "type": "AUDIO" + }, + { + "id": 12, + "origin_id": 10, + "origin_slot": 0, + "target_id": 5, + "target_slot": 2, + "type": "FLOAT" + }, + { + "id": 11, + "origin_id": 3, + "origin_slot": 2, + "target_id": 10, + "target_slot": 0, + "type": "FLOAT" + }, + { + "id": 13, + "origin_id": 11, + "origin_slot": 0, + "target_id": 10, + "target_slot": 1, + "type": "FLOAT" + }, + { + "id": 15, + "origin_id": 13, + "origin_slot": 0, + "target_id": 10, + "target_slot": 2, + "type": "BOOLEAN" + }, + { + "id": 18, + "origin_id": 3, + "origin_slot": 2, + "target_id": 11, + "target_slot": 0, + "type": "FLOAT" + }, + { + "id": 19, + "origin_id": 9, + "origin_slot": 0, + "target_id": 11, + "target_slot": 1, + "type": "INT" + }, + { + "id": 2, + "origin_id": -10, + "origin_slot": 0, + "target_id": 3, + "target_slot": 0, + "type": "VIDEO" + }, + { + "id": 22, + "origin_id": -10, + "origin_slot": 1, + "target_id": 9, + "target_slot": 0, + "type": "INT" + }, + { + "id": 23, + "origin_id": -10, + "origin_slot": 2, + "target_id": 13, + "target_slot": 0, + "type": "BOOLEAN" + }, + { + "id": 24, + "origin_id": -10, + "origin_slot": 3, + "target_id": 1, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 26, + "origin_id": 5, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "VIDEO" + }, + { + "id": 28, + "origin_id": 2, + "origin_slot": 0, + "target_id": -20, + "target_slot": 1, + "type": "IMAGE" + } + ], + "extra": {}, + "category": "Video Tools", + "description": "Increases video frame rate by synthesizing intermediate frames with a frame interpolation model." + } + ] + }, + "extra": {} +} \ No newline at end of file diff --git a/blueprints/Get Any Video Frame.json b/blueprints/Get Any Video Frame.json new file mode 100644 index 000000000..9ff0f8e6e --- /dev/null +++ b/blueprints/Get Any Video Frame.json @@ -0,0 +1,485 @@ +{ + "revision": 0, + "last_node_id": 98, + "last_link_id": 0, + "nodes": [ + { + "id": 98, + "type": "dca6e78d-fb06-421e-97f7-6ce17a665260", + "pos": [ + -410, + -2230 + ], + "size": [ + 270, + 104 + ], + "flags": {}, + "order": 7, + "mode": 0, + "inputs": [ + { + "name": "video", + "type": "VIDEO", + "link": null + }, + { + "label": "frame_index", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": null + } + ], + "outputs": [ + { + "name": "IMAGE", + "type": "IMAGE", + "links": [] + } + ], + "title": "Get Any Video Frame", + "properties": { + "proxyWidgets": [ + [ + "100", + "value" + ] + ] + }, + "widgets_values": [] + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "dca6e78d-fb06-421e-97f7-6ce17a665260", + "version": 1, + "state": { + "lastGroupId": 1, + "lastNodeId": 136, + "lastLinkId": 302, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "Get Any Video Frame", + "inputNode": { + "id": -10, + "bounding": [ + 380, + -57, + 120, + 80 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + 1460, + -57, + 120, + 60 + ] + }, + "inputs": [ + { + "id": "2ceec378-8dcf-4340-8570-155967f59a93", + "name": "video", + "type": "VIDEO", + "linkIds": [ + 4 + ], + "pos": [ + 480, + -37 + ] + }, + { + "id": "819955f6-c686-4896-8032-ff2d0059109a", + "name": "value", + "type": "INT", + "linkIds": [ + 283 + ], + "label": "frame_index", + "pos": [ + 480, + -17 + ] + } + ], + "outputs": [ + { + "id": "1ab0684d-6a44-45b6-8aa4-a0b971a1d41e", + "name": "IMAGE", + "type": "IMAGE", + "linkIds": [ + 5 + ], + "pos": [ + 1480, + -37 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 1, + "type": "GetVideoComponents", + "pos": [ + 560, + -150 + ], + "size": [ + 230, + 120 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "localized_name": "video", + "name": "video", + "type": "VIDEO", + "link": 4 + } + ], + "outputs": [ + { + "localized_name": "images", + "name": "images", + "type": "IMAGE", + "links": [ + 1, + 2 + ] + }, + { + "localized_name": "audio", + "name": "audio", + "type": "AUDIO", + "links": null + }, + { + "localized_name": "fps", + "name": "fps", + "type": "FLOAT", + "links": null + } + ], + "properties": { + "Node name for S&R": "GetVideoComponents" + } + }, + { + "id": 2, + "type": "GetImageSize", + "pos": [ + 560, + 50 + ], + "size": [ + 230, + 120 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 1 + } + ], + "outputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "links": null + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "links": null + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "links": [ + 285 + ] + } + ], + "properties": { + "Node name for S&R": "GetImageSize" + } + }, + { + "id": 3, + "type": "ImageFromBatch", + "pos": [ + 1130, + -150 + ], + "size": [ + 270, + 140 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 2 + }, + { + "localized_name": "batch_index", + "name": "batch_index", + "type": "INT", + "widget": { + "name": "batch_index" + }, + "link": 286 + }, + { + "localized_name": "length", + "name": "length", + "type": "INT", + "widget": { + "name": "length" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [ + 5 + ] + } + ], + "properties": { + "Node name for S&R": "ImageFromBatch" + }, + "widgets_values": [ + 0, + 1 + ] + }, + { + "id": 99, + "type": "ComfyMathExpression", + "pos": [ + 910, + 100 + ], + "size": [ + 400, + 200 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "label": "a", + "localized_name": "values.a", + "name": "values.a", + "type": "FLOAT,INT", + "link": 284 + }, + { + "label": "b", + "localized_name": "values.b", + "name": "values.b", + "shape": 7, + "type": "FLOAT,INT", + "link": 285 + }, + { + "label": "c", + "localized_name": "values.c", + "name": "values.c", + "shape": 7, + "type": "FLOAT,INT", + "link": null + }, + { + "localized_name": "expression", + "name": "expression", + "type": "STRING", + "widget": { + "name": "expression" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "FLOAT", + "name": "FLOAT", + "type": "FLOAT", + "links": null + }, + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 286 + ] + } + ], + "properties": { + "Node name for S&R": "ComfyMathExpression" + }, + "widgets_values": [ + "min(max(int(a if a >= 0 else b + a), 0), b - 1)" + ] + }, + { + "id": 100, + "type": "PrimitiveInt", + "pos": [ + 560, + 250 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": 283 + } + ], + "outputs": [ + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 284 + ] + } + ], + "properties": { + "Node name for S&R": "PrimitiveInt" + }, + "widgets_values": [ + 0, + "fixed" + ] + } + ], + "groups": [], + "links": [ + { + "id": 1, + "origin_id": 1, + "origin_slot": 0, + "target_id": 2, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 2, + "origin_id": 1, + "origin_slot": 0, + "target_id": 3, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 4, + "origin_id": -10, + "origin_slot": 0, + "target_id": 1, + "target_slot": 0, + "type": "VIDEO" + }, + { + "id": 5, + "origin_id": 3, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 283, + "origin_id": -10, + "origin_slot": 1, + "target_id": 100, + "target_slot": 0, + "type": "INT" + }, + { + "id": 284, + "origin_id": 100, + "origin_slot": 0, + "target_id": 99, + "target_slot": 0, + "type": "INT" + }, + { + "id": 285, + "origin_id": 2, + "origin_slot": 2, + "target_id": 99, + "target_slot": 1, + "type": "INT" + }, + { + "id": 286, + "origin_id": 99, + "origin_slot": 1, + "target_id": 3, + "target_slot": 1, + "type": "INT" + } + ], + "extra": {}, + "category": "Video Tools", + "description": "Extracts one image frame from a video at a chosen index, with optional trim and FPS control." + } + ] + }, + "extra": { + "ds": { + "scale": 1.197015527856339, + "offset": [ + -168.76833554248222, + 540.6638955283997 + ] + }, + "frontendVersion": "1.42.8" + } +} \ No newline at end of file diff --git a/blueprints/Image Edit (FireRed Image Edit 1.1).json b/blueprints/Image Edit (FireRed Image Edit 1.1).json index 14310353c..b82c7d18b 100644 --- a/blueprints/Image Edit (FireRed Image Edit 1.1).json +++ b/blueprints/Image Edit (FireRed Image Edit 1.1).json @@ -1,18 +1,18 @@ { "revision": 0, - "last_node_id": 172, + "last_node_id": 213, "last_link_id": 0, "nodes": [ { - "id": 172, - "type": "edf73971-14ee-4d39-b58e-46ce2a89d3d0", + "id": 213, + "type": "e35fbbeb-d7b1-46d1-a74e-959517d0fb1a", "pos": [ - 30, - 200 + -700, + -470 ], "size": [ 500, - 570 + 0 ], "flags": {}, "order": 2, @@ -105,44 +105,44 @@ "properties": { "proxyWidgets": [ [ - "118", + "208", "prompt" ], [ - "153", + "207", "value" ], [ - "130", + "210", "seed" ], [ - "128", + "205", "unet_name" ], [ - "115", + "203", "clip_name" ], [ - "116", + "202", "vae_name" ], [ - "151", + "204", "lora_name" ], [ - "130", + "210", "control_after_generate" ] ], + "cnr_id": "comfy-core", + "ver": "0.15.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.15.1", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -160,12 +160,12 @@ "definitions": { "subgraphs": [ { - "id": "edf73971-14ee-4d39-b58e-46ce2a89d3d0", + "id": "e35fbbeb-d7b1-46d1-a74e-959517d0fb1a", "version": 1, "state": { "lastGroupId": 8, - "lastNodeId": 174, - "lastLinkId": 376, + "lastNodeId": 213, + "lastLinkId": 378, "lastRerouteId": 0 }, "revision": 0, @@ -183,8 +183,8 @@ "outputNode": { "id": -20, "bounding": [ - 1147.5, - -1215, + 1860, + -1340, 120, 60 ] @@ -327,26 +327,26 @@ ], "localized_name": "IMAGE", "pos": [ - 1167.5, - -1195 + 1880, + -1320 ] } ], "widgets": [], "nodes": [ { - "id": 120, + "id": 193, "type": "ModelSamplingAuraFlow", "pos": [ - 1060, - -1760 + 1010, + -1680 ], "size": [ 290, 110 ], "flags": {}, - "order": 8, + "order": 4, "mode": 0, "inputs": [ { @@ -376,13 +376,13 @@ } ], "properties": { + "Node name for S&R": "ModelSamplingAuraFlow", + "cnr_id": "comfy-core", + "ver": "0.5.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.5.1", - "Node name for S&R": "ModelSamplingAuraFlow", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -396,7 +396,7 @@ ] }, { - "id": 154, + "id": 194, "type": "ComfySwitchNode", "pos": [ 680, @@ -407,7 +407,7 @@ 140 ], "flags": {}, - "order": 16, + "order": 5, "mode": 0, "inputs": [ { @@ -444,13 +444,13 @@ ], "title": "Switch (Model)", "properties": { + "Node name for S&R": "ComfySwitchNode", + "cnr_id": "comfy-core", + "ver": "0.15.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.15.1", - "Node name for S&R": "ComfySwitchNode", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -464,7 +464,7 @@ ] }, { - "id": 155, + "id": 195, "type": "PrimitiveInt", "pos": [ 190, @@ -500,13 +500,13 @@ ], "title": "Int (Steps)", "properties": { + "Node name for S&R": "PrimitiveInt", + "cnr_id": "comfy-core", + "ver": "0.15.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.15.1", - "Node name for S&R": "PrimitiveInt", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -521,18 +521,18 @@ ] }, { - "id": 123, + "id": 196, "type": "CFGNorm", "pos": [ - 1060, - -1590 + 1010, + -1510 ], "size": [ 290, 110 ], "flags": {}, - "order": 9, + "order": 6, "mode": 0, "inputs": [ { @@ -562,13 +562,13 @@ } ], "properties": { + "Node name for S&R": "CFGNorm", + "cnr_id": "comfy-core", + "ver": "0.5.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.5.1", - "Node name for S&R": "CFGNorm", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -582,7 +582,7 @@ ] }, { - "id": 164, + "id": 197, "type": "ComfySwitchNode", "pos": [ 680, @@ -593,7 +593,7 @@ 130 ], "flags": {}, - "order": 18, + "order": 7, "mode": 0, "inputs": [ { @@ -630,13 +630,13 @@ ], "title": "Switch (CFG)", "properties": { + "Node name for S&R": "ComfySwitchNode", + "cnr_id": "comfy-core", + "ver": "0.15.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.15.1", - "Node name for S&R": "ComfySwitchNode", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -650,7 +650,7 @@ ] }, { - "id": 156, + "id": 198, "type": "PrimitiveInt", "pos": [ 190, @@ -686,13 +686,13 @@ ], "title": "Float (Steps)", "properties": { + "Node name for S&R": "PrimitiveInt", + "cnr_id": "comfy-core", + "ver": "0.15.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.15.1", - "Node name for S&R": "PrimitiveInt", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -707,7 +707,7 @@ ] }, { - "id": 162, + "id": 199, "type": "PrimitiveFloat", "pos": [ 190, @@ -743,13 +743,13 @@ ], "title": "Float (CFG)", "properties": { + "Node name for S&R": "PrimitiveFloat", + "cnr_id": "comfy-core", + "ver": "0.15.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.15.1", - "Node name for S&R": "PrimitiveFloat", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -763,7 +763,7 @@ ] }, { - "id": 163, + "id": 200, "type": "PrimitiveFloat", "pos": [ 190, @@ -799,13 +799,13 @@ ], "title": "Float (CFG)", "properties": { + "Node name for S&R": "PrimitiveFloat", + "cnr_id": "comfy-core", + "ver": "0.15.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.15.1", - "Node name for S&R": "PrimitiveFloat", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -819,7 +819,7 @@ ] }, { - "id": 157, + "id": 201, "type": "ComfySwitchNode", "pos": [ 680, @@ -830,7 +830,7 @@ 130 ], "flags": {}, - "order": 17, + "order": 8, "mode": 0, "inputs": [ { @@ -867,13 +867,13 @@ ], "title": "Switch (Steps)", "properties": { + "Node name for S&R": "ComfySwitchNode", + "cnr_id": "comfy-core", + "ver": "0.15.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.15.1", - "Node name for S&R": "ComfySwitchNode", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -887,11 +887,11 @@ ] }, { - "id": 116, + "id": 202, "type": "VAELoader", "pos": [ - -950, - -1040 + -960, + -1100 ], "size": [ 400, @@ -900,7 +900,7 @@ "flags": { "collapsed": false }, - "order": 5, + "order": 9, "mode": 0, "inputs": [ { @@ -928,45 +928,45 @@ } ], "properties": { - "ue_properties": { - "widget_ue_connectable": {}, - "input_ue_unconnectable": {} - }, + "Node name for S&R": "VAELoader", "cnr_id": "comfy-core", "ver": "0.5.1", - "Node name for S&R": "VAELoader", - "enableTabs": false, - "tabWidth": 65, - "tabXOffset": 10, - "hasSecondTab": false, - "secondTabText": "Send Back", - "secondTabOffset": 80, - "secondTabWidth": 65, "models": [ { "name": "qwen_image_vae.safetensors", "url": "https://huggingface.co/FireRedTeam/FireRed-Image-Edit-1.0-ComfyUI/resolve/main/qwen_image_vae.safetensors", "directory": "vae" } - ] + ], + "ue_properties": { + "widget_ue_connectable": {}, + "input_ue_unconnectable": {} + }, + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 }, "widgets_values": [ "qwen_image_vae.safetensors" ] }, { - "id": 115, + "id": 203, "type": "CLIPLoader", "pos": [ -960, - -1370 + -1400 ], "size": [ 400, 150 ], "flags": {}, - "order": 4, + "order": 10, "mode": 0, "inputs": [ { @@ -1010,27 +1010,27 @@ } ], "properties": { - "ue_properties": { - "widget_ue_connectable": {}, - "input_ue_unconnectable": {} - }, + "Node name for S&R": "CLIPLoader", "cnr_id": "comfy-core", "ver": "0.5.1", - "Node name for S&R": "CLIPLoader", - "enableTabs": false, - "tabWidth": 65, - "tabXOffset": 10, - "hasSecondTab": false, - "secondTabText": "Send Back", - "secondTabOffset": 80, - "secondTabWidth": 65, "models": [ { "name": "qwen_2.5_vl_7b_fp8_scaled.safetensors", "url": "https://huggingface.co/Comfy-Org/HunyuanVideo_1.5_repackaged/resolve/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors", "directory": "text_encoders" } - ] + ], + "ue_properties": { + "widget_ue_connectable": {}, + "input_ue_unconnectable": {} + }, + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 }, "widgets_values": [ "qwen_2.5_vl_7b_fp8_scaled.safetensors", @@ -1039,7 +1039,7 @@ ] }, { - "id": 151, + "id": 204, "type": "LoraLoaderModelOnly", "pos": [ 100, @@ -1050,7 +1050,7 @@ 140 ], "flags": {}, - "order": 14, + "order": 11, "mode": 0, "inputs": [ { @@ -1089,27 +1089,27 @@ } ], "properties": { - "ue_properties": { - "widget_ue_connectable": {}, - "input_ue_unconnectable": {} - }, + "Node name for S&R": "LoraLoaderModelOnly", "cnr_id": "comfy-core", "ver": "0.15.1", - "Node name for S&R": "LoraLoaderModelOnly", - "enableTabs": false, - "tabWidth": 65, - "tabXOffset": 10, - "hasSecondTab": false, - "secondTabText": "Send Back", - "secondTabOffset": 80, - "secondTabWidth": 65, "models": [ { "name": "FireRed-Image-Edit-1.0-Lightning-8steps-v1.0.safetensors", "url": "https://huggingface.co/FireRedTeam/FireRed-Image-Edit-1.0-ComfyUI/resolve/main/FireRed-Image-Edit-1.0-Lightning-8steps-v1.0.safetensors", "directory": "loras" } - ] + ], + "ue_properties": { + "widget_ue_connectable": {}, + "input_ue_unconnectable": {} + }, + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 }, "widgets_values": [ "FireRed-Image-Edit-1.0-Lightning-8steps-v1.0.safetensors", @@ -1117,7 +1117,7 @@ ] }, { - "id": 128, + "id": 205, "type": "UNETLoader", "pos": [ -960, @@ -1163,27 +1163,27 @@ } ], "properties": { - "ue_properties": { - "widget_ue_connectable": {}, - "input_ue_unconnectable": {} - }, + "Node name for S&R": "UNETLoader", "cnr_id": "comfy-core", "ver": "0.5.1", - "Node name for S&R": "UNETLoader", - "enableTabs": false, - "tabWidth": 65, - "tabXOffset": 10, - "hasSecondTab": false, - "secondTabText": "Send Back", - "secondTabOffset": 80, - "secondTabWidth": 65, "models": [ { "name": "FireRed-Image-Edit-1.1-transformer.safetensors", "url": "https://huggingface.co/FireRedTeam/FireRed-Image-Edit-1.1-ComfyUI/resolve/main/FireRed-Image-Edit-1.1-transformer.safetensors", "directory": "diffusion_models" } - ] + ], + "ue_properties": { + "widget_ue_connectable": {}, + "input_ue_unconnectable": {} + }, + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 }, "widgets_values": [ "FireRed-Image-Edit-1.1-transformer.safetensors", @@ -1191,7 +1191,7 @@ ] }, { - "id": 125, + "id": 206, "type": "VAEEncode", "pos": [ -390, @@ -1202,7 +1202,7 @@ 100 ], "flags": {}, - "order": 10, + "order": 13, "mode": 0, "inputs": [ { @@ -1229,13 +1229,13 @@ } ], "properties": { + "Node name for S&R": "VAEEncode", + "cnr_id": "comfy-core", + "ver": "0.5.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.5.1", - "Node name for S&R": "VAEEncode", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -1246,7 +1246,7 @@ } }, { - "id": 153, + "id": 207, "type": "PrimitiveBoolean", "pos": [ 160, @@ -1257,7 +1257,7 @@ 100 ], "flags": {}, - "order": 15, + "order": 14, "mode": 0, "inputs": [ { @@ -1284,13 +1284,13 @@ ], "title": "Enable Lightning LoRA?", "properties": { + "Node name for S&R": "PrimitiveBoolean", + "cnr_id": "comfy-core", + "ver": "0.15.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.15.1", - "Node name for S&R": "PrimitiveBoolean", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -1304,7 +1304,7 @@ ] }, { - "id": 118, + "id": 208, "type": "TextEncodeQwenImageEditPlus", "pos": [ -480, @@ -1315,7 +1315,7 @@ 370 ], "flags": {}, - "order": 7, + "order": 15, "mode": 0, "inputs": [ { @@ -1374,13 +1374,13 @@ ], "title": "TextEncodeQwenImageEditPlus (Positive)", "properties": { + "Node name for S&R": "TextEncodeQwenImageEditPlus", + "cnr_id": "comfy-core", + "ver": "0.5.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.5.1", - "Node name for S&R": "TextEncodeQwenImageEditPlus", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -1396,7 +1396,7 @@ "bgcolor": "#353" }, { - "id": 117, + "id": 209, "type": "TextEncodeQwenImageEditPlus", "pos": [ -470, @@ -1407,7 +1407,7 @@ 290 ], "flags": {}, - "order": 6, + "order": 16, "mode": 0, "inputs": [ { @@ -1465,13 +1465,13 @@ } ], "properties": { + "Node name for S&R": "TextEncodeQwenImageEditPlus", + "cnr_id": "comfy-core", + "ver": "0.5.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.5.1", - "Node name for S&R": "TextEncodeQwenImageEditPlus", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -1487,18 +1487,18 @@ "bgcolor": "#535" }, { - "id": 130, + "id": 210, "type": "KSampler", "pos": [ - 1060, - -1420 + 1010, + -1340 ], "size": [ 270, 480 ], "flags": {}, - "order": 13, + "order": 17, "mode": 0, "inputs": [ { @@ -1591,13 +1591,13 @@ } ], "properties": { + "Node name for S&R": "KSampler", + "cnr_id": "comfy-core", + "ver": "0.5.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.5.1", - "Node name for S&R": "KSampler", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -1617,11 +1617,11 @@ ] }, { - "id": 126, + "id": 211, "type": "VAEDecode", "pos": [ - 1360, - -1420 + 1440, + -1340 ], "size": [ 230, @@ -1630,7 +1630,7 @@ "flags": { "collapsed": false }, - "order": 11, + "order": 18, "mode": 0, "inputs": [ { @@ -1658,13 +1658,13 @@ } ], "properties": { + "Node name for S&R": "VAEDecode", + "cnr_id": "comfy-core", + "ver": "0.5.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} }, - "cnr_id": "comfy-core", - "ver": "0.5.1", - "Node name for S&R": "VAEDecode", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -1675,7 +1675,7 @@ } }, { - "id": 174, + "id": 212, "type": "ResizeImageMaskNode", "pos": [ -900, @@ -1736,18 +1736,18 @@ } ], "properties": { + "Node name for S&R": "ResizeImageMaskNode", + "cnr_id": "comfy-core", + "ver": "0.18.1", "ue_properties": { "widget_ue_connectable": {}, "input_ue_unconnectable": {} - }, - "cnr_id": "comfy-core", - "ver": "0.18.1", - "Node name for S&R": "ResizeImageMaskNode" + } }, "widgets_values": [ "scale total pixels", 1, - "area" + "lanczos" ] } ], @@ -1808,207 +1808,207 @@ "links": [ { "id": 326, - "origin_id": 154, + "origin_id": 194, "origin_slot": 0, - "target_id": 120, + "target_id": 193, "target_slot": 0, "type": "MODEL" }, { "id": 324, - "origin_id": 128, + "origin_id": 205, "origin_slot": 0, - "target_id": 154, + "target_id": 194, "target_slot": 0, "type": "MODEL" }, { "id": 325, - "origin_id": 151, + "origin_id": 204, "origin_slot": 0, - "target_id": 154, + "target_id": 194, "target_slot": 1, "type": "MODEL" }, { "id": 323, - "origin_id": 153, + "origin_id": 207, "origin_slot": 0, - "target_id": 154, + "target_id": 194, "target_slot": 2, "type": "BOOLEAN" }, { "id": 294, - "origin_id": 120, + "origin_id": 193, "origin_slot": 0, - "target_id": 123, + "target_id": 196, "target_slot": 0, "type": "MODEL" }, { "id": 333, - "origin_id": 162, + "origin_id": 199, "origin_slot": 0, - "target_id": 164, + "target_id": 197, "target_slot": 0, "type": "FLOAT" }, { "id": 334, - "origin_id": 163, + "origin_id": 200, "origin_slot": 0, - "target_id": 164, + "target_id": 197, "target_slot": 1, "type": "FLOAT" }, { "id": 336, - "origin_id": 153, + "origin_id": 207, "origin_slot": 0, - "target_id": 164, + "target_id": 197, "target_slot": 2, "type": "BOOLEAN" }, { "id": 329, - "origin_id": 155, + "origin_id": 195, "origin_slot": 0, - "target_id": 157, + "target_id": 201, "target_slot": 0, "type": "INT" }, { "id": 337, - "origin_id": 156, + "origin_id": 198, "origin_slot": 0, - "target_id": 157, + "target_id": 201, "target_slot": 1, "type": "INT" }, { "id": 330, - "origin_id": 153, + "origin_id": 207, "origin_slot": 0, - "target_id": 157, + "target_id": 201, "target_slot": 2, "type": "BOOLEAN" }, { "id": 297, - "origin_id": 115, + "origin_id": 203, "origin_slot": 0, - "target_id": 117, + "target_id": 209, "target_slot": 0, "type": "CLIP" }, { "id": 299, - "origin_id": 116, + "origin_id": 202, "origin_slot": 0, - "target_id": 117, + "target_id": 209, "target_slot": 1, "type": "VAE" }, { "id": 316, - "origin_id": 128, + "origin_id": 205, "origin_slot": 0, - "target_id": 151, + "target_id": 204, "target_slot": 0, "type": "MODEL" }, { "id": 296, - "origin_id": 115, + "origin_id": 203, "origin_slot": 0, - "target_id": 118, + "target_id": 208, "target_slot": 0, "type": "CLIP" }, { "id": 298, - "origin_id": 116, + "origin_id": 202, "origin_slot": 0, - "target_id": 118, + "target_id": 208, "target_slot": 1, "type": "VAE" }, { "id": 300, - "origin_id": 116, + "origin_id": 202, "origin_slot": 0, - "target_id": 125, + "target_id": 206, "target_slot": 1, "type": "VAE" }, { "id": 295, - "origin_id": 123, + "origin_id": 196, "origin_slot": 0, - "target_id": 130, + "target_id": 210, "target_slot": 0, "type": "MODEL" }, { "id": 312, - "origin_id": 118, + "origin_id": 208, "origin_slot": 0, - "target_id": 130, + "target_id": 210, "target_slot": 1, "type": "CONDITIONING" }, { "id": 313, - "origin_id": 117, + "origin_id": 209, "origin_slot": 0, - "target_id": 130, + "target_id": 210, "target_slot": 2, "type": "CONDITIONING" }, { "id": 303, - "origin_id": 125, + "origin_id": 206, "origin_slot": 0, - "target_id": 130, + "target_id": 210, "target_slot": 3, "type": "LATENT" }, { "id": 345, - "origin_id": 157, + "origin_id": 201, "origin_slot": 0, - "target_id": 130, + "target_id": 210, "target_slot": 5, "type": "INT" }, { "id": 335, - "origin_id": 164, + "origin_id": 197, "origin_slot": 0, - "target_id": 130, + "target_id": 210, "target_slot": 6, "type": "FLOAT" }, { "id": 273, - "origin_id": 130, + "origin_id": 210, "origin_slot": 0, - "target_id": 126, + "target_id": 211, "target_slot": 0, "type": "LATENT" }, { "id": 314, - "origin_id": 116, + "origin_id": 202, "origin_slot": 0, - "target_id": 126, + "target_id": 211, "target_slot": 1, "type": "VAE" }, { "id": 292, - "origin_id": 126, + "origin_id": 211, "origin_slot": 0, "target_id": -20, "target_slot": 0, @@ -2018,7 +2018,7 @@ "id": 355, "origin_id": -10, "origin_slot": 1, - "target_id": 118, + "target_id": 208, "target_slot": 3, "type": "IMAGE" }, @@ -2026,7 +2026,7 @@ "id": 356, "origin_id": -10, "origin_slot": 1, - "target_id": 117, + "target_id": 209, "target_slot": 3, "type": "IMAGE" }, @@ -2034,7 +2034,7 @@ "id": 357, "origin_id": -10, "origin_slot": 2, - "target_id": 118, + "target_id": 208, "target_slot": 4, "type": "IMAGE" }, @@ -2042,7 +2042,7 @@ "id": 358, "origin_id": -10, "origin_slot": 2, - "target_id": 117, + "target_id": 209, "target_slot": 4, "type": "IMAGE" }, @@ -2050,7 +2050,7 @@ "id": 359, "origin_id": -10, "origin_slot": 3, - "target_id": 118, + "target_id": 208, "target_slot": 5, "type": "STRING" }, @@ -2058,31 +2058,31 @@ "id": 364, "origin_id": -10, "origin_slot": 4, - "target_id": 153, + "target_id": 207, "target_slot": 0, "type": "BOOLEAN" }, { "id": 368, - "origin_id": 174, + "origin_id": 212, "origin_slot": 0, - "target_id": 125, + "target_id": 206, "target_slot": 0, "type": "IMAGE" }, { "id": 369, - "origin_id": 174, + "origin_id": 212, "origin_slot": 0, - "target_id": 118, + "target_id": 208, "target_slot": 2, "type": "IMAGE" }, { "id": 370, - "origin_id": 174, + "origin_id": 212, "origin_slot": 0, - "target_id": 117, + "target_id": 209, "target_slot": 2, "type": "IMAGE" }, @@ -2090,7 +2090,7 @@ "id": 371, "origin_id": -10, "origin_slot": 0, - "target_id": 174, + "target_id": 212, "target_slot": 0, "type": "IMAGE" }, @@ -2098,7 +2098,7 @@ "id": 372, "origin_id": -10, "origin_slot": 5, - "target_id": 130, + "target_id": 210, "target_slot": 4, "type": "INT" }, @@ -2106,7 +2106,7 @@ "id": 373, "origin_id": -10, "origin_slot": 6, - "target_id": 128, + "target_id": 205, "target_slot": 0, "type": "COMBO" }, @@ -2114,7 +2114,7 @@ "id": 374, "origin_id": -10, "origin_slot": 7, - "target_id": 115, + "target_id": 203, "target_slot": 0, "type": "COMBO" }, @@ -2122,7 +2122,7 @@ "id": 375, "origin_id": -10, "origin_slot": 8, - "target_id": 116, + "target_id": 202, "target_slot": 0, "type": "COMBO" }, @@ -2130,7 +2130,7 @@ "id": 376, "origin_id": -10, "origin_slot": 9, - "target_id": 151, + "target_id": 204, "target_slot": 1, "type": "COMBO" } diff --git a/blueprints/Image Edit (Flux.2 Dev).json b/blueprints/Image Edit (Flux.2 Dev).json new file mode 100644 index 000000000..92827bf17 --- /dev/null +++ b/blueprints/Image Edit (Flux.2 Dev).json @@ -0,0 +1,2050 @@ +{ + "revision": 0, + "last_node_id": 139, + "last_link_id": 0, + "nodes": [ + { + "id": 139, + "type": "41b0c117-7470-454c-914e-b8742dc06d62", + "pos": [ + -650, + 570 + ], + "size": [ + 400, + 0 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "label": "image", + "localized_name": "pixels", + "name": "pixels", + "type": "IMAGE", + "link": null + }, + { + "label": "prompt", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": null + }, + { + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": null + }, + { + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": null + }, + { + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": null + }, + { + "label": "enable_turbo_mode", + "name": "value", + "type": "BOOLEAN", + "widget": { + "name": "value" + }, + "link": null + }, + { + "label": "turbo_lora", + "name": "lora_name", + "type": "COMBO", + "widget": { + "name": "lora_name" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [] + } + ], + "properties": { + "proxyWidgets": [ + [ + "123", + "text" + ], + [ + "129", + "unet_name" + ], + [ + "124", + "clip_name" + ], + [ + "121", + "vae_name" + ], + [ + "138", + "value" + ], + [ + "128", + "lora_name" + ], + [ + "125", + "noise_seed" + ], + [ + "125", + "control_after_generate" + ] + ], + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": { + "text": true, + "value": true, + "lora_name": true + }, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [], + "title": "Image Edit (Flux.2 Dev)" + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "41b0c117-7470-454c-914e-b8742dc06d62", + "version": 1, + "state": { + "lastGroupId": 8, + "lastNodeId": 139, + "lastLinkId": 194, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "Image Edit (Flux.2 Dev)", + "inputNode": { + "id": -10, + "bounding": [ + -1520, + 400, + 151.744140625, + 180 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + 1240, + 420, + 120, + 60 + ] + }, + "inputs": [ + { + "id": "fc74acd5-30a9-410b-abb5-4a4171ba3d25", + "name": "pixels", + "type": "IMAGE", + "linkIds": [ + 126, + 169 + ], + "localized_name": "pixels", + "label": "image", + "pos": [ + -1388.255859375, + 420 + ] + }, + { + "id": "3e69affa-397b-4d52-82d7-68dfcef9e761", + "name": "text", + "type": "STRING", + "linkIds": [ + 168 + ], + "label": "prompt", + "pos": [ + -1388.255859375, + 440 + ] + }, + { + "id": "2f016a8a-fb3e-4cb9-97f2-a991defe4fa2", + "name": "unet_name", + "type": "COMBO", + "linkIds": [ + 177 + ], + "pos": [ + -1388.255859375, + 460 + ] + }, + { + "id": "799b9dc7-0c90-4b19-9a13-e01d896bea1f", + "name": "clip_name", + "type": "COMBO", + "linkIds": [ + 178 + ], + "pos": [ + -1388.255859375, + 480 + ] + }, + { + "id": "e58a83c9-1b93-4378-9598-f24068820313", + "name": "vae_name", + "type": "COMBO", + "linkIds": [ + 179 + ], + "pos": [ + -1388.255859375, + 500 + ] + }, + { + "id": "8335a4a9-0ce4-4e67-a641-1c9d7a762977", + "name": "value", + "type": "BOOLEAN", + "linkIds": [ + 191 + ], + "label": "enable_turbo_mode", + "pos": [ + -1388.255859375, + 520 + ] + }, + { + "id": "890b22b4-44a7-4707-912a-ca8b4ee7b7c9", + "name": "lora_name", + "type": "COMBO", + "linkIds": [ + 192 + ], + "label": "turbo_lora", + "pos": [ + -1388.255859375, + 540 + ] + } + ], + "outputs": [ + { + "id": "3eaa05d6-4960-4a7c-bf2a-8b585fbb7c9c", + "name": "IMAGE", + "type": "IMAGE", + "linkIds": [ + 9 + ], + "localized_name": "IMAGE", + "pos": [ + 1260, + 440 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 118, + "type": "Flux2Scheduler", + "pos": [ + 540, + 430 + ], + "size": [ + 230, + 170 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [ + { + "localized_name": "steps", + "name": "steps", + "type": "INT", + "widget": { + "name": "steps" + }, + "link": 188 + }, + { + "localized_name": "width", + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": 170 + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": 172 + } + ], + "outputs": [ + { + "localized_name": "SIGMAS", + "name": "SIGMAS", + "type": "SIGMAS", + "links": [ + 132 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "Flux2Scheduler", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 20, + 1248, + 832 + ] + }, + { + "id": 119, + "type": "BasicGuider", + "pos": [ + 530, + 120 + ], + "size": [ + 230, + 100 + ], + "flags": {}, + "order": 5, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 185 + }, + { + "localized_name": "conditioning", + "name": "conditioning", + "type": "CONDITIONING", + "link": 166 + } + ], + "outputs": [ + { + "localized_name": "GUIDER", + "name": "GUIDER", + "type": "GUIDER", + "slot_index": 0, + "links": [ + 30 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "BasicGuider", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 120, + "type": "KSamplerSelect", + "pos": [ + 530, + 270 + ], + "size": [ + 230, + 110 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "localized_name": "sampler_name", + "name": "sampler_name", + "type": "COMBO", + "widget": { + "name": "sampler_name" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "SAMPLER", + "name": "SAMPLER", + "type": "SAMPLER", + "links": [ + 19 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "KSamplerSelect", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "euler" + ] + }, + { + "id": 121, + "type": "VAELoader", + "pos": [ + -970, + 390 + ], + "size": [ + 300, + 110 + ], + "flags": {}, + "order": 6, + "mode": 0, + "inputs": [ + { + "localized_name": "vae_name", + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": 179 + } + ], + "outputs": [ + { + "localized_name": "VAE", + "name": "VAE", + "type": "VAE", + "slot_index": 0, + "links": [ + 127, + 159 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "VAELoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "full_encoder_small_decoder.safetensors", + "url": "https://huggingface.co/black-forest-labs/FLUX.2-small-decoder/resolve/main/full_encoder_small_decoder.safetensors", + "directory": "vae" + } + ] + }, + "widgets_values": [ + "full_encoder_small_decoder.safetensors" + ] + }, + { + "id": 122, + "type": "SamplerCustomAdvanced", + "pos": [ + 790, + -50 + ], + "size": [ + 280, + 170 + ], + "flags": {}, + "order": 7, + "mode": 0, + "inputs": [ + { + "localized_name": "noise", + "name": "noise", + "type": "NOISE", + "link": 37 + }, + { + "localized_name": "guider", + "name": "guider", + "type": "GUIDER", + "link": 30 + }, + { + "localized_name": "sampler", + "name": "sampler", + "type": "SAMPLER", + "link": 19 + }, + { + "localized_name": "sigmas", + "name": "sigmas", + "type": "SIGMAS", + "link": 132 + }, + { + "localized_name": "latent_image", + "name": "latent_image", + "type": "LATENT", + "link": 161 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "LATENT", + "slot_index": 0, + "links": [ + 24 + ] + }, + { + "localized_name": "denoised_output", + "name": "denoised_output", + "type": "LATENT", + "links": null + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "SamplerCustomAdvanced", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 123, + "type": "CLIPTextEncode", + "pos": [ + -630, + -50 + ], + "size": [ + 430, + 360 + ], + "flags": {}, + "order": 8, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 117 + }, + { + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": 168 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "slot_index": 0, + "links": [ + 41 + ] + } + ], + "title": "CLIP Text Encode (Positive Prompt)", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "CLIPTextEncode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "" + ], + "color": "#232", + "bgcolor": "#353" + }, + { + "id": 124, + "type": "CLIPLoader", + "pos": [ + -970, + 160 + ], + "size": [ + 300, + 150 + ], + "flags": {}, + "order": 9, + "mode": 0, + "inputs": [ + { + "localized_name": "clip_name", + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": 178 + }, + { + "localized_name": "type", + "name": "type", + "type": "COMBO", + "widget": { + "name": "type" + }, + "link": null + }, + { + "localized_name": "device", + "name": "device", + "shape": 7, + "type": "COMBO", + "widget": { + "name": "device" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [ + 117 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "CLIPLoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "mistral_3_small_flux2_bf16.safetensors", + "url": "https://huggingface.co/Comfy-Org/flux2-dev/resolve/main/split_files/text_encoders/mistral_3_small_flux2_bf16.safetensors", + "directory": "text_encoders" + } + ] + }, + "widgets_values": [ + "mistral_3_small_flux2_bf16.safetensors", + "flux2", + "default" + ] + }, + { + "id": 125, + "type": "RandomNoise", + "pos": [ + 530, + -50 + ], + "size": [ + 230, + 110 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "noise_seed", + "name": "noise_seed", + "type": "INT", + "widget": { + "name": "noise_seed" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "NOISE", + "name": "NOISE", + "type": "NOISE", + "links": [ + 37 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "RandomNoise", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 342971778941390, + "randomize" + ] + }, + { + "id": 126, + "type": "VAEDecode", + "pos": [ + 830, + 410 + ], + "size": [ + 230, + 100 + ], + "flags": {}, + "order": 10, + "mode": 0, + "inputs": [ + { + "localized_name": "samples", + "name": "samples", + "type": "LATENT", + "link": 24 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 159 + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "slot_index": 0, + "links": [ + 9 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "VAEDecode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 127, + "type": "FluxGuidance", + "pos": [ + -520, + 390 + ], + "size": [ + 320, + 110 + ], + "flags": {}, + "order": 11, + "mode": 0, + "inputs": [ + { + "localized_name": "conditioning", + "name": "conditioning", + "type": "CONDITIONING", + "link": 41 + }, + { + "localized_name": "guidance", + "name": "guidance", + "type": "FLOAT", + "widget": { + "name": "guidance" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "slot_index": 0, + "links": [ + 144 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "FluxGuidance", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 4 + ], + "color": "#233", + "bgcolor": "#355" + }, + { + "id": 128, + "type": "LoraLoaderModelOnly", + "pos": [ + -150, + 200 + ], + "size": [ + 300, + 140 + ], + "flags": {}, + "order": 12, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 181 + }, + { + "localized_name": "lora_name", + "name": "lora_name", + "type": "COMBO", + "widget": { + "name": "lora_name" + }, + "link": 192 + }, + { + "localized_name": "strength_model", + "name": "strength_model", + "type": "FLOAT", + "widget": { + "name": "strength_model" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 183 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "LoraLoaderModelOnly", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "Flux_2-Turbo-LoRA_comfyui.safetensors", + "url": "https://huggingface.co/ByteZSzn/Flux.2-Turbo-ComfyUI/resolve/main/Flux_2-Turbo-LoRA_comfyui.safetensors", + "directory": "loras" + } + ] + }, + "widgets_values": [ + "Flux_2-Turbo-LoRA_comfyui.safetensors", + 1 + ] + }, + { + "id": 129, + "type": "UNETLoader", + "pos": [ + -970, + -40 + ], + "size": [ + 300, + 110 + ], + "flags": {}, + "order": 13, + "mode": 0, + "inputs": [ + { + "localized_name": "unet_name", + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": 177 + }, + { + "localized_name": "weight_dtype", + "name": "weight_dtype", + "type": "COMBO", + "widget": { + "name": "weight_dtype" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "slot_index": 0, + "links": [ + 181, + 184 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "UNETLoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "flux2_dev_fp8mixed.safetensors", + "url": "https://huggingface.co/Comfy-Org/flux2-dev/resolve/main/split_files/diffusion_models/flux2_dev_fp8mixed.safetensors", + "directory": "diffusion_models" + } + ] + }, + "widgets_values": [ + "flux2_dev_fp8mixed.safetensors", + "default" + ] + }, + { + "id": 130, + "type": "ComfySwitchNode", + "pos": [ + 220, + 10 + ], + "size": [ + 270, + 130 + ], + "flags": {}, + "order": 14, + "mode": 0, + "inputs": [ + { + "localized_name": "on_false", + "name": "on_false", + "type": "*", + "link": 184 + }, + { + "localized_name": "on_true", + "name": "on_true", + "type": "*", + "link": 183 + }, + { + "localized_name": "switch", + "name": "switch", + "type": "BOOLEAN", + "widget": { + "name": "switch" + }, + "link": 190 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "*", + "links": [ + 185 + ] + } + ], + "title": "Switch(model)", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.18.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ComfySwitchNode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + false + ] + }, + { + "id": 131, + "type": "PrimitiveInt", + "pos": [ + -150, + 430 + ], + "size": [ + 300, + 110 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 186 + ] + } + ], + "title": "Steps", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.18.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "PrimitiveInt", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 8, + "fixed" + ] + }, + { + "id": 132, + "type": "PrimitiveInt", + "pos": [ + -150, + -50 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 187 + ] + } + ], + "title": "Steps", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.18.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "PrimitiveInt", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 20, + "fixed" + ] + }, + { + "id": 133, + "type": "ComfySwitchNode", + "pos": [ + 220, + 280 + ], + "size": [ + 270, + 130 + ], + "flags": {}, + "order": 15, + "mode": 0, + "inputs": [ + { + "localized_name": "on_false", + "name": "on_false", + "type": "*", + "link": 187 + }, + { + "localized_name": "on_true", + "name": "on_true", + "type": "*", + "link": 186 + }, + { + "localized_name": "switch", + "name": "switch", + "type": "BOOLEAN", + "widget": { + "name": "switch" + }, + "link": 189 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "*", + "links": [ + 188 + ] + } + ], + "title": "Switch(steps)", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.18.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ComfySwitchNode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + false + ] + }, + { + "id": 134, + "type": "EmptyFlux2LatentImage", + "pos": [ + 530, + 790 + ], + "size": [ + 270, + 170 + ], + "flags": {}, + "order": 16, + "mode": 0, + "inputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": 171 + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": 173 + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "widget": { + "name": "batch_size" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "links": [ + 161 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "EmptyFlux2LatentImage", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 1248, + 832, + 1 + ] + }, + { + "id": 135, + "type": "GetImageSize", + "pos": [ + -100, + 810 + ], + "size": [ + 230, + 120 + ], + "flags": {}, + "order": 17, + "mode": 0, + "inputs": [ + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 169 + } + ], + "outputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "links": [ + 170, + 171 + ] + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "links": [ + 172, + 173 + ] + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "links": null + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "GetImageSize", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 136, + "type": "VAEEncode", + "pos": [ + -910, + 600 + ], + "size": [ + 230, + 100 + ], + "flags": { + "collapsed": true + }, + "order": 18, + "mode": 0, + "inputs": [ + { + "localized_name": "pixels", + "name": "pixels", + "type": "IMAGE", + "link": 126 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 127 + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "links": [ + 125 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "VAEEncode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 137, + "type": "ReferenceLatent", + "pos": [ + -470, + 580 + ], + "size": [ + 230, + 100 + ], + "flags": {}, + "order": 19, + "mode": 0, + "inputs": [ + { + "localized_name": "conditioning", + "name": "conditioning", + "type": "CONDITIONING", + "link": 144 + }, + { + "localized_name": "latent", + "name": "latent", + "shape": 7, + "type": "LATENT", + "link": 125 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 166 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ReferenceLatent", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 138, + "type": "PrimitiveBoolean", + "pos": [ + -130, + 640 + ], + "size": [ + 270, + 100 + ], + "flags": {}, + "order": 20, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "BOOLEAN", + "widget": { + "name": "value" + }, + "link": 191 + } + ], + "outputs": [ + { + "localized_name": "BOOLEAN", + "name": "BOOLEAN", + "type": "BOOLEAN", + "links": [ + 189, + 190 + ] + } + ], + "title": "Enable 8 steps lora", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.18.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "PrimitiveBoolean", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + false + ] + } + ], + "groups": [ + { + "id": 1, + "title": "Models", + "bounding": [ + -980, + -120, + 320, + 640 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 2, + "title": "Custom sampler", + "bounding": [ + 520, + -120, + 590, + 740 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 3, + "title": "Image size", + "bounding": [ + 510, + 690, + 590, + 290 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 4, + "title": "Prompt", + "bounding": [ + -640, + -120, + 450, + 640 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 7, + "title": "Original", + "bounding": [ + -160, + -120, + 340, + 230 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 8, + "title": "8 Steps LoRA", + "bounding": [ + -160, + 130, + 340, + 430 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + } + ], + "links": [ + { + "id": 41, + "origin_id": 123, + "origin_slot": 0, + "target_id": 127, + "target_slot": 0, + "type": "CONDITIONING" + }, + { + "id": 144, + "origin_id": 127, + "origin_slot": 0, + "target_id": 137, + "target_slot": 0, + "type": "CONDITIONING" + }, + { + "id": 125, + "origin_id": 136, + "origin_slot": 0, + "target_id": 137, + "target_slot": 1, + "type": "LATENT" + }, + { + "id": 37, + "origin_id": 125, + "origin_slot": 0, + "target_id": 122, + "target_slot": 0, + "type": "NOISE" + }, + { + "id": 30, + "origin_id": 119, + "origin_slot": 0, + "target_id": 122, + "target_slot": 1, + "type": "GUIDER" + }, + { + "id": 19, + "origin_id": 120, + "origin_slot": 0, + "target_id": 122, + "target_slot": 2, + "type": "SAMPLER" + }, + { + "id": 132, + "origin_id": 118, + "origin_slot": 0, + "target_id": 122, + "target_slot": 3, + "type": "SIGMAS" + }, + { + "id": 161, + "origin_id": 134, + "origin_slot": 0, + "target_id": 122, + "target_slot": 4, + "type": "LATENT" + }, + { + "id": 24, + "origin_id": 122, + "origin_slot": 0, + "target_id": 126, + "target_slot": 0, + "type": "LATENT" + }, + { + "id": 159, + "origin_id": 121, + "origin_slot": 0, + "target_id": 126, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 117, + "origin_id": 124, + "origin_slot": 0, + "target_id": 123, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 127, + "origin_id": 121, + "origin_slot": 0, + "target_id": 136, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 126, + "origin_id": -10, + "origin_slot": 0, + "target_id": 136, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 9, + "origin_id": 126, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 166, + "origin_id": 137, + "origin_slot": 0, + "target_id": 119, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 168, + "origin_id": -10, + "origin_slot": 1, + "target_id": 123, + "target_slot": 1, + "type": "STRING" + }, + { + "id": 169, + "origin_id": -10, + "origin_slot": 0, + "target_id": 135, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 170, + "origin_id": 135, + "origin_slot": 0, + "target_id": 118, + "target_slot": 1, + "type": "INT" + }, + { + "id": 171, + "origin_id": 135, + "origin_slot": 0, + "target_id": 134, + "target_slot": 0, + "type": "INT" + }, + { + "id": 172, + "origin_id": 135, + "origin_slot": 1, + "target_id": 118, + "target_slot": 2, + "type": "INT" + }, + { + "id": 173, + "origin_id": 135, + "origin_slot": 1, + "target_id": 134, + "target_slot": 1, + "type": "INT" + }, + { + "id": 177, + "origin_id": -10, + "origin_slot": 2, + "target_id": 129, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 178, + "origin_id": -10, + "origin_slot": 3, + "target_id": 124, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 179, + "origin_id": -10, + "origin_slot": 4, + "target_id": 121, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 181, + "origin_id": 129, + "origin_slot": 0, + "target_id": 128, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 183, + "origin_id": 128, + "origin_slot": 0, + "target_id": 130, + "target_slot": 1, + "type": "MODEL" + }, + { + "id": 184, + "origin_id": 129, + "origin_slot": 0, + "target_id": 130, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 185, + "origin_id": 130, + "origin_slot": 0, + "target_id": 119, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 186, + "origin_id": 131, + "origin_slot": 0, + "target_id": 133, + "target_slot": 1, + "type": "INT" + }, + { + "id": 187, + "origin_id": 132, + "origin_slot": 0, + "target_id": 133, + "target_slot": 0, + "type": "INT" + }, + { + "id": 188, + "origin_id": 133, + "origin_slot": 0, + "target_id": 118, + "target_slot": 0, + "type": "INT" + }, + { + "id": 189, + "origin_id": 138, + "origin_slot": 0, + "target_id": 133, + "target_slot": 2, + "type": "BOOLEAN" + }, + { + "id": 190, + "origin_id": 138, + "origin_slot": 0, + "target_id": 130, + "target_slot": 2, + "type": "BOOLEAN" + }, + { + "id": 191, + "origin_id": -10, + "origin_slot": 5, + "target_id": 138, + "target_slot": 0, + "type": "BOOLEAN" + }, + { + "id": 192, + "origin_id": -10, + "origin_slot": 6, + "target_id": 128, + "target_slot": 1, + "type": "COMBO" + } + ], + "extra": { + "workflowRendererVersion": "LG" + }, + "category": "Image generation and editing/Edit image", + "description": "Edits an image from text instructions using Flux.2 [dev], with guidance, schedulers, and optional Turbo LoRAs." + } + ] + }, + "extra": { + "ue_links": [] + } +} \ No newline at end of file diff --git a/blueprints/Image Edit (Qwen 2509).json b/blueprints/Image Edit (Qwen 2509).json new file mode 100644 index 000000000..f7be322a0 --- /dev/null +++ b/blueprints/Image Edit (Qwen 2509).json @@ -0,0 +1,1947 @@ +{ + "revision": 0, + "last_node_id": 433, + "last_link_id": 0, + "nodes": [ + { + "id": 433, + "type": "eba40a3a-f6c5-48ac-b58e-55525d06b373", + "pos": [ + 90, + -160 + ], + "size": [ + 390, + 610 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "label": "image", + "name": "image", + "type": "IMAGE", + "link": null + }, + { + "label": "image2 (optional)", + "name": "image2", + "type": "IMAGE", + "link": null + }, + { + "label": "image3 (optional)", + "name": "image3", + "type": "IMAGE", + "link": null + }, + { + "name": "prompt", + "type": "STRING", + "widget": { + "name": "prompt" + }, + "link": null + }, + { + "name": "seed", + "type": "INT", + "widget": { + "name": "seed" + }, + "link": null + }, + { + "label": "enable_turbo_mode", + "name": "value", + "type": "BOOLEAN", + "widget": { + "name": "value" + }, + "link": null + }, + { + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": null + }, + { + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": null + }, + { + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [] + } + ], + "properties": { + "proxyWidgets": [ + [ + "111", + "prompt" + ], + [ + "3", + "seed" + ], + [ + "443", + "value" + ], + [ + "37", + "unet_name" + ], + [ + "38", + "clip_name" + ], + [ + "39", + "vae_name" + ], + [ + "3", + "control_after_generate" + ] + ], + "cnr_id": "comfy-core", + "ver": "0.3.62" + }, + "widgets_values": [], + "title": "Image Edit (Qwen 2509)" + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "eba40a3a-f6c5-48ac-b58e-55525d06b373", + "version": 1, + "state": { + "lastGroupId": 51, + "lastNodeId": 468, + "lastLinkId": 731, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "Image Edit (Qwen 2509)", + "inputNode": { + "id": -10, + "bounding": [ + -1160, + 280, + 151.744140625, + 220 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + 2030, + -20, + 120, + 60 + ] + }, + "inputs": [ + { + "id": "d5089bd3-63bc-4a24-b478-6565ed2364e3", + "name": "image", + "type": "IMAGE", + "linkIds": [ + 248 + ], + "label": "image", + "pos": [ + -1028.255859375, + 300 + ] + }, + { + "id": "9e80fff0-ed0a-439f-a16e-a4a6cc1eb601", + "name": "image2", + "type": "IMAGE", + "linkIds": [ + 235, + 236 + ], + "label": "image2 (optional)", + "pos": [ + -1028.255859375, + 320 + ] + }, + { + "id": "49d98fd6-01b5-440b-8603-579252fd7fef", + "name": "image3", + "type": "IMAGE", + "linkIds": [ + 237, + 238 + ], + "label": "image3 (optional)", + "pos": [ + -1028.255859375, + 340 + ] + }, + { + "id": "5de32f24-a7b5-4423-b772-72824005f585", + "name": "prompt", + "type": "STRING", + "linkIds": [ + 244 + ], + "pos": [ + -1028.255859375, + 360 + ] + }, + { + "id": "85fb3d74-7881-4c71-bc8c-624be5eedc3d", + "name": "seed", + "type": "INT", + "linkIds": [ + 718 + ], + "pos": [ + -1028.255859375, + 380 + ] + }, + { + "id": "b0c828de-d7eb-42a3-8dfb-4f53360d4fc9", + "name": "value", + "type": "BOOLEAN", + "linkIds": [ + 719 + ], + "label": "enable_turbo_mode", + "pos": [ + -1028.255859375, + 400 + ] + }, + { + "id": "072baa05-5551-4a98-bd66-015a36833ac2", + "name": "unet_name", + "type": "COMBO", + "linkIds": [ + 720 + ], + "pos": [ + -1028.255859375, + 420 + ] + }, + { + "id": "d2891d11-b336-4750-9742-b93717c9ae39", + "name": "clip_name", + "type": "COMBO", + "linkIds": [ + 721 + ], + "pos": [ + -1028.255859375, + 440 + ] + }, + { + "id": "4218135f-5128-4b7e-8572-92cc55615793", + "name": "vae_name", + "type": "COMBO", + "linkIds": [ + 722 + ], + "pos": [ + -1028.255859375, + 460 + ] + } + ], + "outputs": [ + { + "id": "c4ebfc18-de83-4361-8e42-767c3c8c25c0", + "name": "IMAGE", + "type": "IMAGE", + "linkIds": [ + 110 + ], + "localized_name": "IMAGE", + "pos": [ + 2050, + 0 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 75, + "type": "CFGNorm", + "pos": [ + 1080, + 30 + ], + "size": [ + 290, + 110 + ], + "flags": {}, + "order": 11, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 141 + }, + { + "localized_name": "strength", + "name": "strength", + "type": "FLOAT", + "widget": { + "name": "strength" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "patched_model", + "name": "patched_model", + "type": "MODEL", + "links": [ + 186 + ] + } + ], + "properties": { + "Node name for S&R": "CFGNorm", + "cnr_id": "comfy-core", + "ver": "0.3.50", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": { + "strength": true + } + } + }, + "widgets_values": [ + 1 + ] + }, + { + "id": 39, + "type": "VAELoader", + "pos": [ + -730, + 410 + ], + "size": [ + 330, + 110 + ], + "flags": {}, + "order": 9, + "mode": 0, + "inputs": [ + { + "localized_name": "vae_name", + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": 722 + } + ], + "outputs": [ + { + "localized_name": "VAE", + "name": "VAE", + "type": "VAE", + "slot_index": 0, + "links": [ + 76, + 168, + 206, + 207 + ] + } + ], + "properties": { + "Node name for S&R": "VAELoader", + "cnr_id": "comfy-core", + "ver": "0.3.48", + "models": [ + { + "name": "qwen_image_vae.safetensors", + "url": "https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/vae/qwen_image_vae.safetensors", + "directory": "vae" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "widget_ue_connectable": {} + }, + "widgets_values": [ + "qwen_image_vae.safetensors" + ] + }, + { + "id": 38, + "type": "CLIPLoader", + "pos": [ + -730, + 150 + ], + "size": [ + 330, + 150 + ], + "flags": {}, + "order": 8, + "mode": 0, + "inputs": [ + { + "localized_name": "clip_name", + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": 721 + }, + { + "localized_name": "type", + "name": "type", + "type": "COMBO", + "widget": { + "name": "type" + }, + "link": null + }, + { + "localized_name": "device", + "name": "device", + "shape": 7, + "type": "COMBO", + "widget": { + "name": "device" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "slot_index": 0, + "links": [ + 204, + 205 + ] + } + ], + "properties": { + "Node name for S&R": "CLIPLoader", + "cnr_id": "comfy-core", + "ver": "0.3.48", + "models": [ + { + "name": "qwen_2.5_vl_7b_fp8_scaled.safetensors", + "url": "https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors", + "directory": "text_encoders" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "widget_ue_connectable": {} + }, + "widgets_values": [ + "qwen_2.5_vl_7b_fp8_scaled.safetensors", + "qwen_image", + "default" + ] + }, + { + "id": 37, + "type": "UNETLoader", + "pos": [ + -730, + -60 + ], + "size": [ + 330, + 110 + ], + "flags": {}, + "order": 7, + "mode": 0, + "inputs": [ + { + "localized_name": "unet_name", + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": 720 + }, + { + "localized_name": "weight_dtype", + "name": "weight_dtype", + "type": "COMBO", + "widget": { + "name": "weight_dtype" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "slot_index": 0, + "links": [ + 184, + 710 + ] + } + ], + "properties": { + "Node name for S&R": "UNETLoader", + "cnr_id": "comfy-core", + "ver": "0.3.48", + "models": [ + { + "name": "qwen_image_edit_2509_fp8_e4m3fn.safetensors", + "url": "https://huggingface.co/Comfy-Org/Qwen-Image-Edit_ComfyUI/resolve/main/split_files/diffusion_models/qwen_image_edit_2509_fp8_e4m3fn.safetensors", + "directory": "diffusion_models" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "widget_ue_connectable": {} + }, + "widgets_values": [ + "qwen_image_edit_2509_fp8_e4m3fn.safetensors", + "default" + ] + }, + { + "id": 110, + "type": "TextEncodeQwenImageEditPlus", + "pos": [ + -240, + 320 + ], + "size": [ + 400, + 240 + ], + "flags": {}, + "order": 14, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 204 + }, + { + "localized_name": "vae", + "name": "vae", + "shape": 7, + "type": "VAE", + "link": 206 + }, + { + "localized_name": "image1", + "name": "image1", + "shape": 7, + "type": "IMAGE", + "link": 251 + }, + { + "localized_name": "image2", + "name": "image2", + "shape": 7, + "type": "IMAGE", + "link": 236 + }, + { + "localized_name": "image3", + "name": "image3", + "shape": 7, + "type": "IMAGE", + "link": 238 + }, + { + "localized_name": "prompt", + "name": "prompt", + "type": "STRING", + "widget": { + "name": "prompt" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 210 + ] + } + ], + "properties": { + "Node name for S&R": "TextEncodeQwenImageEditPlus", + "cnr_id": "comfy-core", + "ver": "0.3.59" + }, + "widgets_values": [ + "" + ], + "color": "#223", + "bgcolor": "#335" + }, + { + "id": 66, + "type": "ModelSamplingAuraFlow", + "pos": [ + 1070, + -120 + ], + "size": [ + 290, + 110 + ], + "flags": {}, + "order": 10, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 708 + }, + { + "localized_name": "shift", + "name": "shift", + "type": "FLOAT", + "widget": { + "name": "shift" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 141 + ] + } + ], + "properties": { + "Node name for S&R": "ModelSamplingAuraFlow", + "cnr_id": "comfy-core", + "ver": "0.3.48", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "widget_ue_connectable": {} + }, + "widgets_values": [ + 3 + ] + }, + { + "id": 111, + "type": "TextEncodeQwenImageEditPlus", + "pos": [ + -250, + -70 + ], + "size": [ + 410, + 330 + ], + "flags": {}, + "order": 15, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 205 + }, + { + "localized_name": "vae", + "name": "vae", + "shape": 7, + "type": "VAE", + "link": 207 + }, + { + "localized_name": "image1", + "name": "image1", + "shape": 7, + "type": "IMAGE", + "link": 250 + }, + { + "localized_name": "image2", + "name": "image2", + "shape": 7, + "type": "IMAGE", + "link": 235 + }, + { + "localized_name": "image3", + "name": "image3", + "shape": 7, + "type": "IMAGE", + "link": 237 + }, + { + "localized_name": "prompt", + "name": "prompt", + "type": "STRING", + "widget": { + "name": "prompt" + }, + "link": 244 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 211 + ] + } + ], + "properties": { + "Node name for S&R": "TextEncodeQwenImageEditPlus", + "cnr_id": "comfy-core", + "ver": "0.3.59" + }, + "widgets_values": [ + "" + ], + "color": "#232", + "bgcolor": "#353" + }, + { + "id": 88, + "type": "VAEEncode", + "pos": [ + -70, + 640 + ], + "size": [ + 230, + 100 + ], + "flags": {}, + "order": 12, + "mode": 0, + "inputs": [ + { + "localized_name": "pixels", + "name": "pixels", + "type": "IMAGE", + "link": 249 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 168 + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "links": [ + 246 + ] + } + ], + "properties": { + "Node name for S&R": "VAEEncode", + "cnr_id": "comfy-core", + "ver": "0.3.50", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {} + } + } + }, + { + "id": 8, + "type": "VAEDecode", + "pos": [ + 1590, + -60 + ], + "size": [ + 230, + 100 + ], + "flags": { + "collapsed": false + }, + "order": 6, + "mode": 0, + "inputs": [ + { + "localized_name": "samples", + "name": "samples", + "type": "LATENT", + "link": 128 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 76 + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "slot_index": 0, + "links": [ + 110 + ] + } + ], + "properties": { + "Node name for S&R": "VAEDecode", + "cnr_id": "comfy-core", + "ver": "0.3.48", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "widget_ue_connectable": {} + } + }, + { + "id": 89, + "type": "LoraLoaderModelOnly", + "pos": [ + 320, + 300 + ], + "size": [ + 300, + 140 + ], + "flags": {}, + "order": 13, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 184 + }, + { + "localized_name": "lora_name", + "name": "lora_name", + "type": "COMBO", + "widget": { + "name": "lora_name" + }, + "link": null + }, + { + "localized_name": "strength_model", + "name": "strength_model", + "type": "FLOAT", + "widget": { + "name": "strength_model" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 709 + ] + } + ], + "properties": { + "Node name for S&R": "LoraLoaderModelOnly", + "cnr_id": "comfy-core", + "ver": "0.3.50", + "models": [ + { + "name": "Qwen-Image-Edit-2509-Lightning-4steps-V1.0-bf16.safetensors", + "url": "https://huggingface.co/lightx2v/Qwen-Image-Lightning/resolve/main/Qwen-Image-Edit-2509/Qwen-Image-Edit-2509-Lightning-4steps-V1.0-bf16.safetensors", + "directory": "loras" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": { + "lora_name": true, + "strength_model": true + } + } + }, + "widgets_values": [ + "Qwen-Image-Edit-2509-Lightning-4steps-V1.0-bf16.safetensors", + 1 + ] + }, + { + "id": 117, + "type": "FluxKontextImageScale", + "pos": [ + -680, + 630 + ], + "size": [ + 230, + 80 + ], + "flags": {}, + "order": 16, + "mode": 0, + "inputs": [ + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 248 + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [ + 249, + 250, + 251 + ] + } + ], + "properties": { + "Node name for S&R": "FluxKontextImageScale" + } + }, + { + "id": 3, + "type": "KSampler", + "pos": [ + 1070, + 210 + ], + "size": [ + 300, + 590 + ], + "flags": {}, + "order": 5, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 186 + }, + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "link": 211 + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "link": 210 + }, + { + "localized_name": "latent_image", + "name": "latent_image", + "type": "LATENT", + "link": 246 + }, + { + "localized_name": "seed", + "name": "seed", + "type": "INT", + "widget": { + "name": "seed" + }, + "link": 718 + }, + { + "localized_name": "steps", + "name": "steps", + "type": "INT", + "widget": { + "name": "steps" + }, + "link": 707 + }, + { + "localized_name": "cfg", + "name": "cfg", + "type": "FLOAT", + "widget": { + "name": "cfg" + }, + "link": 706 + }, + { + "localized_name": "sampler_name", + "name": "sampler_name", + "type": "COMBO", + "widget": { + "name": "sampler_name" + }, + "link": null + }, + { + "localized_name": "scheduler", + "name": "scheduler", + "type": "COMBO", + "widget": { + "name": "scheduler" + }, + "link": null + }, + { + "localized_name": "denoise", + "name": "denoise", + "type": "FLOAT", + "widget": { + "name": "denoise" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "slot_index": 0, + "links": [ + 128 + ] + } + ], + "properties": { + "Node name for S&R": "KSampler", + "cnr_id": "comfy-core", + "ver": "0.3.48", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "widget_ue_connectable": {} + }, + "widgets_values": [ + 973414316252139, + "randomize", + 4, + 1, + "euler", + "simple", + 1 + ] + }, + { + "id": 436, + "type": "PrimitiveInt", + "pos": [ + 320, + 500 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 713 + ] + } + ], + "title": "Steps", + "properties": { + "Node name for S&R": "PrimitiveInt" + }, + "widgets_values": [ + 4, + "fixed" + ] + }, + { + "id": 437, + "type": "PrimitiveFloat", + "pos": [ + 320, + 670 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "FLOAT", + "widget": { + "name": "value" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "FLOAT", + "name": "FLOAT", + "type": "FLOAT", + "links": [ + 714 + ] + } + ], + "title": "CFG", + "properties": { + "Node name for S&R": "PrimitiveFloat" + }, + "widgets_values": [ + 1 + ] + }, + { + "id": 438, + "type": "PrimitiveInt", + "pos": [ + 320, + -100 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 711 + ] + } + ], + "title": "Steps", + "properties": { + "Node name for S&R": "PrimitiveInt" + }, + "widgets_values": [ + 20, + "fixed" + ] + }, + { + "id": 439, + "type": "PrimitiveFloat", + "pos": [ + 320, + 70 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "FLOAT", + "widget": { + "name": "value" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "FLOAT", + "name": "FLOAT", + "type": "FLOAT", + "links": [ + 712 + ] + } + ], + "title": "CFG", + "properties": { + "Node name for S&R": "PrimitiveFloat" + }, + "widgets_values": [ + 4 + ] + }, + { + "id": 440, + "type": "ComfySwitchNode", + "pos": [ + 750, + -80 + ], + "size": [ + 270, + 130 + ], + "flags": {}, + "order": 17, + "mode": 0, + "inputs": [ + { + "localized_name": "on_false", + "name": "on_false", + "type": "*", + "link": 710 + }, + { + "localized_name": "on_true", + "name": "on_true", + "type": "*", + "link": 709 + }, + { + "localized_name": "switch", + "name": "switch", + "type": "BOOLEAN", + "widget": { + "name": "switch" + }, + "link": 715 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "*", + "links": [ + 708 + ] + } + ], + "title": "Switch (Model)", + "properties": { + "Node name for S&R": "ComfySwitchNode" + }, + "widgets_values": [ + false + ] + }, + { + "id": 441, + "type": "ComfySwitchNode", + "pos": [ + 730, + 340 + ], + "size": [ + 270, + 130 + ], + "flags": {}, + "order": 18, + "mode": 0, + "inputs": [ + { + "localized_name": "on_false", + "name": "on_false", + "type": "*", + "link": 711 + }, + { + "localized_name": "on_true", + "name": "on_true", + "type": "*", + "link": 713 + }, + { + "localized_name": "switch", + "name": "switch", + "type": "BOOLEAN", + "widget": { + "name": "switch" + }, + "link": 716 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "*", + "links": [ + 707 + ] + } + ], + "title": "Switch (Steps)", + "properties": { + "Node name for S&R": "ComfySwitchNode" + }, + "widgets_values": [ + false + ] + }, + { + "id": 442, + "type": "ComfySwitchNode", + "pos": [ + 730, + 520 + ], + "size": [ + 270, + 130 + ], + "flags": {}, + "order": 19, + "mode": 0, + "inputs": [ + { + "localized_name": "on_false", + "name": "on_false", + "type": "*", + "link": 712 + }, + { + "localized_name": "on_true", + "name": "on_true", + "type": "*", + "link": 714 + }, + { + "localized_name": "switch", + "name": "switch", + "type": "BOOLEAN", + "widget": { + "name": "switch" + }, + "link": 717 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "*", + "links": [ + 706 + ] + } + ], + "title": "Switch (CFG)", + "properties": { + "Node name for S&R": "ComfySwitchNode" + }, + "widgets_values": [ + false + ] + }, + { + "id": 443, + "type": "PrimitiveBoolean", + "pos": [ + 330, + 850 + ], + "size": [ + 270, + 100 + ], + "flags": {}, + "order": 20, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "BOOLEAN", + "widget": { + "name": "value" + }, + "link": 719 + } + ], + "outputs": [ + { + "localized_name": "BOOLEAN", + "name": "BOOLEAN", + "type": "BOOLEAN", + "links": [ + 715, + 716, + 717 + ] + } + ], + "title": "Enable Lightning LoRA", + "properties": { + "Node name for S&R": "PrimitiveBoolean" + }, + "widgets_values": [ + true + ] + }, + { + "id": 444, + "type": "MarkdownNote", + "pos": [ + 240, + -500 + ], + "size": [ + 450, + 310 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [], + "outputs": [], + "title": "Note: KSampler settings", + "properties": {}, + "widgets_values": [ + "You can test and find the best setting by yourself. The following table is for reference.\n| Parameters | Qwen Team | Comfy Original | with 4steps LoRA |\n|--------|---------|------------|---------------------------|\n| Steps | 50 | 20 | 4 |\n| CFG | 4.0 | 2.5 | 1.0 |" + ], + "color": "#432", + "bgcolor": "#000" + } + ], + "groups": [ + { + "id": 1, + "title": "Step1 - Load models", + "bounding": [ + -770, + -170, + 410, + 750 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 3, + "title": "Step 4 - Prompt", + "bounding": [ + -330, + -170, + 570, + 750 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 50, + "title": "Lightning LoRA", + "bounding": [ + 270, + 220, + 390, + 570 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 51, + "title": "Original Settings", + "bounding": [ + 270, + -170, + 390, + 360 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + } + ], + "links": [ + { + "id": 141, + "origin_id": 66, + "origin_slot": 0, + "target_id": 75, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 128, + "origin_id": 3, + "origin_slot": 0, + "target_id": 8, + "target_slot": 0, + "type": "LATENT" + }, + { + "id": 76, + "origin_id": 39, + "origin_slot": 0, + "target_id": 8, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 184, + "origin_id": 37, + "origin_slot": 0, + "target_id": 89, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 186, + "origin_id": 75, + "origin_slot": 0, + "target_id": 3, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 211, + "origin_id": 111, + "origin_slot": 0, + "target_id": 3, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 210, + "origin_id": 110, + "origin_slot": 0, + "target_id": 3, + "target_slot": 2, + "type": "CONDITIONING" + }, + { + "id": 168, + "origin_id": 39, + "origin_slot": 0, + "target_id": 88, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 204, + "origin_id": 38, + "origin_slot": 0, + "target_id": 110, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 206, + "origin_id": 39, + "origin_slot": 0, + "target_id": 110, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 205, + "origin_id": 38, + "origin_slot": 0, + "target_id": 111, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 207, + "origin_id": 39, + "origin_slot": 0, + "target_id": 111, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 110, + "origin_id": 8, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 235, + "origin_id": -10, + "origin_slot": 1, + "target_id": 111, + "target_slot": 3, + "type": "IMAGE" + }, + { + "id": 236, + "origin_id": -10, + "origin_slot": 1, + "target_id": 110, + "target_slot": 3, + "type": "IMAGE" + }, + { + "id": 237, + "origin_id": -10, + "origin_slot": 2, + "target_id": 111, + "target_slot": 4, + "type": "IMAGE" + }, + { + "id": 238, + "origin_id": -10, + "origin_slot": 2, + "target_id": 110, + "target_slot": 4, + "type": "IMAGE" + }, + { + "id": 244, + "origin_id": -10, + "origin_slot": 3, + "target_id": 111, + "target_slot": 5, + "type": "STRING" + }, + { + "id": 246, + "origin_id": 88, + "origin_slot": 0, + "target_id": 3, + "target_slot": 3, + "type": "LATENT" + }, + { + "id": 248, + "origin_id": -10, + "origin_slot": 0, + "target_id": 117, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 249, + "origin_id": 117, + "origin_slot": 0, + "target_id": 88, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 250, + "origin_id": 117, + "origin_slot": 0, + "target_id": 111, + "target_slot": 2, + "type": "IMAGE" + }, + { + "id": 251, + "origin_id": 117, + "origin_slot": 0, + "target_id": 110, + "target_slot": 2, + "type": "IMAGE" + }, + { + "id": 706, + "origin_id": 442, + "origin_slot": 0, + "target_id": 3, + "target_slot": 6, + "type": "FLOAT" + }, + { + "id": 707, + "origin_id": 441, + "origin_slot": 0, + "target_id": 3, + "target_slot": 5, + "type": "INT" + }, + { + "id": 708, + "origin_id": 440, + "origin_slot": 0, + "target_id": 66, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 709, + "origin_id": 89, + "origin_slot": 0, + "target_id": 440, + "target_slot": 1, + "type": "MODEL" + }, + { + "id": 710, + "origin_id": 37, + "origin_slot": 0, + "target_id": 440, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 711, + "origin_id": 438, + "origin_slot": 0, + "target_id": 441, + "target_slot": 0, + "type": "INT" + }, + { + "id": 712, + "origin_id": 439, + "origin_slot": 0, + "target_id": 442, + "target_slot": 0, + "type": "FLOAT" + }, + { + "id": 713, + "origin_id": 436, + "origin_slot": 0, + "target_id": 441, + "target_slot": 1, + "type": "INT" + }, + { + "id": 714, + "origin_id": 437, + "origin_slot": 0, + "target_id": 442, + "target_slot": 1, + "type": "FLOAT" + }, + { + "id": 715, + "origin_id": 443, + "origin_slot": 0, + "target_id": 440, + "target_slot": 2, + "type": "BOOLEAN" + }, + { + "id": 716, + "origin_id": 443, + "origin_slot": 0, + "target_id": 441, + "target_slot": 2, + "type": "BOOLEAN" + }, + { + "id": 717, + "origin_id": 443, + "origin_slot": 0, + "target_id": 442, + "target_slot": 2, + "type": "BOOLEAN" + }, + { + "id": 718, + "origin_id": -10, + "origin_slot": 4, + "target_id": 3, + "target_slot": 4, + "type": "INT" + }, + { + "id": 719, + "origin_id": -10, + "origin_slot": 5, + "target_id": 443, + "target_slot": 0, + "type": "BOOLEAN" + }, + { + "id": 720, + "origin_id": -10, + "origin_slot": 6, + "target_id": 37, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 721, + "origin_id": -10, + "origin_slot": 7, + "target_id": 38, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 722, + "origin_id": -10, + "origin_slot": 8, + "target_id": 39, + "target_slot": 0, + "type": "COMBO" + } + ], + "extra": { + "workflowRendererVersion": "LG" + }, + "category": "Image generation and editing/Edit image", + "description": "Edits images from text instructions using Qwen-Image-Edit-2509 with optional Lightning LoRA for few-step sampling." + } + ] + }, + "extra": {} +} diff --git a/blueprints/Image Segmentation (SAM3).json b/blueprints/Image Segmentation (SAM3).json new file mode 100644 index 000000000..b405bf623 --- /dev/null +++ b/blueprints/Image Segmentation (SAM3).json @@ -0,0 +1,714 @@ +{ + "revision": 0, + "last_node_id": 99, + "last_link_id": 0, + "nodes": [ + { + "id": 99, + "type": "6e7ab3ea-96aa-470f-9b94-3d9d0e01f481", + "pos": [ + -1630, + -3270 + ], + "size": [ + 290, + 370 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "label": "image", + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": null + }, + { + "label": "object", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": null + }, + { + "name": "bboxes", + "type": "BOUNDING_BOX", + "link": null + }, + { + "name": "positive_coords", + "type": "STRING", + "link": null + }, + { + "name": "negative_coords", + "type": "STRING", + "link": null + }, + { + "name": "threshold", + "type": "FLOAT", + "widget": { + "name": "threshold" + }, + "link": null + }, + { + "name": "refine_iterations", + "type": "INT", + "widget": { + "name": "refine_iterations" + }, + "link": null + }, + { + "name": "individual_masks", + "type": "BOOLEAN", + "widget": { + "name": "individual_masks" + }, + "link": null + }, + { + "name": "ckpt_name", + "type": "COMBO", + "widget": { + "name": "ckpt_name" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "masks", + "name": "masks", + "type": "MASK", + "links": [] + }, + { + "localized_name": "bboxes", + "name": "bboxes", + "type": "BOUNDING_BOX", + "links": [] + } + ], + "properties": { + "proxyWidgets": [ + [ + "78", + "text" + ], + [ + "75", + "threshold" + ], + [ + "75", + "refine_iterations" + ], + [ + "75", + "individual_masks" + ], + [ + "77", + "ckpt_name" + ] + ], + "ue_properties": { + "widget_ue_connectable": { + "text": true + }, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "cnr_id": "comfy-core", + "ver": "0.19.3", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [], + "title": "Image Segmentation (SAM3)" + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "6e7ab3ea-96aa-470f-9b94-3d9d0e01f481", + "version": 1, + "state": { + "lastGroupId": 0, + "lastNodeId": 113, + "lastLinkId": 283, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "Image Segmentation (SAM3)", + "inputNode": { + "id": -10, + "bounding": [ + -2260, + -3450, + 136.369140625, + 220 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + -1130, + -3305, + 120, + 80 + ] + }, + "inputs": [ + { + "id": "a6e75fa2-162a-4af0-a2fd-1e9c899a5ab6", + "name": "image", + "type": "IMAGE", + "linkIds": [ + 264 + ], + "localized_name": "image", + "label": "image", + "pos": [ + -2143.630859375, + -3430 + ] + }, + { + "id": "3cefd304-7631-4ff6-a5a0-5a0ffb120745", + "name": "text", + "type": "STRING", + "linkIds": [ + 265 + ], + "label": "object", + "pos": [ + -2143.630859375, + -3410 + ] + }, + { + "id": "1aec91c5-d8d2-441c-928c-49c14e7e80ed", + "name": "bboxes", + "type": "BOUNDING_BOX", + "linkIds": [ + 266 + ], + "pos": [ + -2143.630859375, + -3390 + ] + }, + { + "id": "1ec7ce1a-8257-4719-8a81-60ebc8a98899", + "name": "positive_coords", + "type": "STRING", + "linkIds": [ + 267 + ], + "pos": [ + -2143.630859375, + -3370 + ] + }, + { + "id": "c65f8b87-9bd7-48be-9fc2-823431e95019", + "name": "negative_coords", + "type": "STRING", + "linkIds": [ + 268 + ], + "pos": [ + -2143.630859375, + -3350 + ] + }, + { + "id": "bb4ba35a-ccfe-4c37-98e5-d9b0d69585fb", + "name": "threshold", + "type": "FLOAT", + "linkIds": [ + 269 + ], + "pos": [ + -2143.630859375, + -3330 + ] + }, + { + "id": "b1439668-b050-490b-a5dc-fc4052c55666", + "name": "refine_iterations", + "type": "INT", + "linkIds": [ + 270 + ], + "pos": [ + -2143.630859375, + -3310 + ] + }, + { + "id": "86e239e5-c098-4302-b54d-d42a38bc0f89", + "name": "individual_masks", + "type": "BOOLEAN", + "linkIds": [ + 271 + ], + "pos": [ + -2143.630859375, + -3290 + ] + }, + { + "id": "f9e0b9d4-b2f1-4907-a4a5-305656576706", + "name": "ckpt_name", + "type": "COMBO", + "linkIds": [ + 272 + ], + "pos": [ + -2143.630859375, + -3270 + ] + } + ], + "outputs": [ + { + "id": "ff50da09-1e59-4a58-9b7f-be1a00aa5913", + "name": "masks", + "type": "MASK", + "linkIds": [ + 231 + ], + "localized_name": "masks", + "pos": [ + -1110, + -3285 + ] + }, + { + "id": "8f622e40-8528-4078-b7d3-147e9f872194", + "name": "bboxes", + "type": "BOUNDING_BOX", + "linkIds": [ + 232 + ], + "localized_name": "bboxes", + "pos": [ + -1110, + -3265 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 75, + "type": "SAM3_Detect", + "pos": [ + -1470, + -3460 + ], + "size": [ + 270, + 260 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "label": "model", + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 237 + }, + { + "label": "image", + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 264 + }, + { + "label": "conditioning", + "localized_name": "conditioning", + "name": "conditioning", + "shape": 7, + "type": "CONDITIONING", + "link": 200 + }, + { + "label": "bboxes", + "localized_name": "bboxes", + "name": "bboxes", + "shape": 7, + "type": "BOUNDING_BOX", + "link": 266 + }, + { + "label": "positive_coords", + "localized_name": "positive_coords", + "name": "positive_coords", + "shape": 7, + "type": "STRING", + "link": 267 + }, + { + "label": "negative_coords", + "localized_name": "negative_coords", + "name": "negative_coords", + "shape": 7, + "type": "STRING", + "link": 268 + }, + { + "localized_name": "threshold", + "name": "threshold", + "type": "FLOAT", + "widget": { + "name": "threshold" + }, + "link": 269 + }, + { + "localized_name": "refine_iterations", + "name": "refine_iterations", + "type": "INT", + "widget": { + "name": "refine_iterations" + }, + "link": 270 + }, + { + "localized_name": "individual_masks", + "name": "individual_masks", + "type": "BOOLEAN", + "widget": { + "name": "individual_masks" + }, + "link": 271 + } + ], + "outputs": [ + { + "localized_name": "masks", + "name": "masks", + "type": "MASK", + "links": [ + 231 + ] + }, + { + "localized_name": "bboxes", + "name": "bboxes", + "type": "BOUNDING_BOX", + "links": [ + 232 + ] + } + ], + "properties": { + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "cnr_id": "comfy-core", + "ver": "0.19.3", + "Node name for S&R": "SAM3_Detect", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 0.5, + 2, + false + ] + }, + { + "id": 77, + "type": "CheckpointLoaderSimple", + "pos": [ + -1970, + -3200 + ], + "size": [ + 330, + 140 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "ckpt_name", + "name": "ckpt_name", + "type": "COMBO", + "widget": { + "name": "ckpt_name" + }, + "link": 272 + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 237 + ] + }, + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [ + 240 + ] + }, + { + "localized_name": "VAE", + "name": "VAE", + "type": "VAE", + "links": null + } + ], + "properties": { + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "cnr_id": "comfy-core", + "ver": "0.19.3", + "Node name for S&R": "CheckpointLoaderSimple", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "sam3.1_multiplex_fp16.safetensors", + "url": "https://huggingface.co/Comfy-Org/sam3.1/resolve/main/checkpoints/sam3.1_multiplex_fp16.safetensors", + "directory": "checkpoints" + } + ] + }, + "widgets_values": [ + "sam3.1_multiplex_fp16.safetensors" + ] + }, + { + "id": 78, + "type": "CLIPTextEncode", + "pos": [ + -2000, + -3000 + ], + "size": [ + 400, + 200 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 240 + }, + { + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": 265 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 200 + ] + } + ], + "properties": { + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "cnr_id": "comfy-core", + "ver": "0.19.3", + "Node name for S&R": "CLIPTextEncode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "" + ] + } + ], + "groups": [], + "links": [ + { + "id": 237, + "origin_id": 77, + "origin_slot": 0, + "target_id": 75, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 200, + "origin_id": 78, + "origin_slot": 0, + "target_id": 75, + "target_slot": 2, + "type": "CONDITIONING" + }, + { + "id": 240, + "origin_id": 77, + "origin_slot": 1, + "target_id": 78, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 231, + "origin_id": 75, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "MASK" + }, + { + "id": 232, + "origin_id": 75, + "origin_slot": 1, + "target_id": -20, + "target_slot": 1, + "type": "BOUNDING_BOX" + }, + { + "id": 264, + "origin_id": -10, + "origin_slot": 0, + "target_id": 75, + "target_slot": 1, + "type": "IMAGE" + }, + { + "id": 265, + "origin_id": -10, + "origin_slot": 1, + "target_id": 78, + "target_slot": 1, + "type": "STRING" + }, + { + "id": 266, + "origin_id": -10, + "origin_slot": 2, + "target_id": 75, + "target_slot": 3, + "type": "BOUNDING_BOX" + }, + { + "id": 267, + "origin_id": -10, + "origin_slot": 3, + "target_id": 75, + "target_slot": 4, + "type": "STRING" + }, + { + "id": 268, + "origin_id": -10, + "origin_slot": 4, + "target_id": 75, + "target_slot": 5, + "type": "STRING" + }, + { + "id": 269, + "origin_id": -10, + "origin_slot": 5, + "target_id": 75, + "target_slot": 6, + "type": "FLOAT" + }, + { + "id": 270, + "origin_id": -10, + "origin_slot": 6, + "target_id": 75, + "target_slot": 7, + "type": "INT" + }, + { + "id": 271, + "origin_id": -10, + "origin_slot": 7, + "target_id": 75, + "target_slot": 8, + "type": "BOOLEAN" + }, + { + "id": 272, + "origin_id": -10, + "origin_slot": 8, + "target_id": 77, + "target_slot": 0, + "type": "COMBO" + } + ], + "extra": {}, + "category": "Image Tools/Image Segmentation", + "description": "Segments images into masks using Meta SAM3 from text prompts, points, or boxes." + } + ] + }, + "extra": { + "ue_links": [] + } +} diff --git a/blueprints/Image to Video (Wan 2.2).json b/blueprints/Image to Video (Wan 2.2).json index 3510aad18..a24adcfb6 100644 --- a/blueprints/Image to Video (Wan 2.2).json +++ b/blueprints/Image to Video (Wan 2.2).json @@ -2028,7 +2028,7 @@ "workflowRendererVersion": "LG" }, "category": "Video generation and editing/Image to video", - "description": "Generates video from an image and text prompt using Wan 2.2, supporting T2V and I2V." + "description": "Image-to-video with Wan 2.2 using a start image plus text prompt to extend motion from the still frame." } ] }, diff --git a/blueprints/Remove Background (BiRefNet).json b/blueprints/Remove Background (BiRefNet).json new file mode 100644 index 000000000..732a4adc4 --- /dev/null +++ b/blueprints/Remove Background (BiRefNet).json @@ -0,0 +1,397 @@ +{ + "revision": 0, + "last_node_id": 19, + "last_link_id": 0, + "nodes": [ + { + "id": 19, + "type": "5b40ca21-ba1a-41d5-b403-4d2d7acdc195", + "pos": [ + -6411.330578108367, + 1940.2638932730042 + ], + "size": [ + 349.609375, + 145.9375 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": null + }, + { + "name": "bg_removal_name", + "type": "COMBO", + "widget": { + "name": "bg_removal_name" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [] + }, + { + "name": "mask", + "type": "MASK", + "links": [] + } + ], + "properties": { + "proxyWidgets": [ + [ + "14", + "bg_removal_name" + ] + ] + }, + "widgets_values": [], + "title": "Remove Background (BiRefNet)" + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "5b40ca21-ba1a-41d5-b403-4d2d7acdc195", + "version": 1, + "state": { + "lastGroupId": 0, + "lastNodeId": 21, + "lastLinkId": 16, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "Remove Background (BiRefNet)", + "description": "Removes or replaces image backgrounds using BiRefNet segmentation and alpha compositing.", + "inputNode": { + "id": -10, + "bounding": [ + -6728.534070722246, + 1475.2619799128663, + 150.9140625, + 88 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + -6169.049695722246, + 1475.2619799128663, + 128, + 88 + ] + }, + "inputs": [ + { + "id": "7bc321cd-df31-4c39-aaf7-7f0d01326189", + "name": "image", + "type": "IMAGE", + "linkIds": [ + 5, + 7 + ], + "localized_name": "image", + "pos": [ + -6601.620008222246, + 1499.2619799128663 + ] + }, + { + "id": "e89d2cd8-daa3-4e29-8a69-851db85072cb", + "name": "bg_removal_name", + "type": "COMBO", + "linkIds": [ + 12 + ], + "pos": [ + -6601.620008222246, + 1519.2619799128663 + ] + } + ], + "outputs": [ + { + "id": "16e7863c-4c38-46c2-aa74-e82991fbfe8d", + "name": "IMAGE", + "type": "IMAGE", + "linkIds": [ + 8 + ], + "localized_name": "IMAGE", + "pos": [ + -6145.049695722246, + 1499.2619799128663 + ] + }, + { + "id": "f7240c19-5b80-406e-a8e2-9b12440ee2d6", + "name": "mask", + "type": "MASK", + "linkIds": [ + 11 + ], + "pos": [ + -6145.049695722246, + 1519.2619799128663 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 13, + "type": "RemoveBackground", + "pos": [ + -6536.764823982709, + 1444.9963409012412 + ], + "size": [ + 302.25, + 72 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 5 + }, + { + "localized_name": "bg_removal_model", + "name": "bg_removal_model", + "type": "BACKGROUND_REMOVAL", + "link": 3 + } + ], + "outputs": [ + { + "localized_name": "mask", + "name": "mask", + "type": "MASK", + "links": [ + 4, + 11 + ] + } + ], + "properties": { + "Node name for S&R": "RemoveBackground" + } + }, + { + "id": 14, + "type": "LoadBackgroundRemovalModel", + "pos": [ + -6540.534070722246, + 1302.223464635445 + ], + "size": [ + 311.484375, + 85.515625 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "bg_removal_name", + "name": "bg_removal_name", + "type": "COMBO", + "widget": { + "name": "bg_removal_name" + }, + "link": 12 + } + ], + "outputs": [ + { + "localized_name": "bg_model", + "name": "bg_model", + "type": "BACKGROUND_REMOVAL", + "links": [ + 3 + ] + } + ], + "properties": { + "Node name for S&R": "LoadBackgroundRemovalModel", + "models": [ + { + "name": "birefnet.safetensors", + "url": "https://huggingface.co/Comfy-Org/BiRefNet/resolve/main/background_removal/birefnet.safetensors", + "directory": "background_removal" + } + ] + }, + "widgets_values": [ + "birefnet.safetensors" + ] + }, + { + "id": 15, + "type": "InvertMask", + "pos": [ + -6532.446160529669, + 1571.1111286839914 + ], + "size": [ + 285.984375, + 48 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "mask", + "name": "mask", + "type": "MASK", + "link": 4 + } + ], + "outputs": [ + { + "localized_name": "MASK", + "name": "MASK", + "type": "MASK", + "links": [ + 6 + ] + } + ], + "properties": { + "Node name for S&R": "InvertMask" + } + }, + { + "id": 16, + "type": "JoinImageWithAlpha", + "pos": [ + -6527.4370171636665, + 1674.3004951902876 + ], + "size": [ + 284.96875, + 72 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 7 + }, + { + "localized_name": "alpha", + "name": "alpha", + "type": "MASK", + "link": 6 + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [ + 8 + ] + } + ], + "properties": { + "Node name for S&R": "JoinImageWithAlpha" + } + } + ], + "groups": [], + "links": [ + { + "id": 3, + "origin_id": 14, + "origin_slot": 0, + "target_id": 13, + "target_slot": 1, + "type": "BACKGROUND_REMOVAL" + }, + { + "id": 4, + "origin_id": 13, + "origin_slot": 0, + "target_id": 15, + "target_slot": 0, + "type": "MASK" + }, + { + "id": 6, + "origin_id": 15, + "origin_slot": 0, + "target_id": 16, + "target_slot": 1, + "type": "MASK" + }, + { + "id": 5, + "origin_id": -10, + "origin_slot": 0, + "target_id": 13, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 7, + "origin_id": -10, + "origin_slot": 0, + "target_id": 16, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 8, + "origin_id": 16, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 11, + "origin_id": 13, + "origin_slot": 0, + "target_id": -20, + "target_slot": 1, + "type": "MASK" + }, + { + "id": 12, + "origin_id": -10, + "origin_slot": 1, + "target_id": 14, + "target_slot": 0, + "type": "COMBO" + } + ], + "extra": {}, + "category": "Image generation and editing/Background Removal" + } + ] + }, + "extra": {} +} \ No newline at end of file diff --git a/blueprints/Text to Image (Ernie Image Turbo).json b/blueprints/Text to Image (Ernie Image Turbo).json new file mode 100644 index 000000000..4ecdd1883 --- /dev/null +++ b/blueprints/Text to Image (Ernie Image Turbo).json @@ -0,0 +1,2112 @@ +{ + "revision": 0, + "last_node_id": 88, + "last_link_id": 0, + "nodes": [ + { + "id": 88, + "type": "2a4f0815-c4d2-4e8b-9bdf-991a8403889d", + "pos": [ + -120, + 240 + ], + "size": [ + 400, + 540 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "label": "prompt", + "name": "value", + "type": "STRING", + "widget": { + "name": "value" + }, + "link": null + }, + { + "label": "prompt_enhancement", + "name": "value_1", + "type": "BOOLEAN", + "widget": { + "name": "value_1" + }, + "link": null + }, + { + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": null + }, + { + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": null + }, + { + "name": "seed", + "type": "INT", + "widget": { + "name": "seed" + }, + "link": null + }, + { + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": null + }, + { + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": null + }, + { + "label": "prompt_enhancer", + "name": "clip_name_1", + "type": "COMBO", + "widget": { + "name": "clip_name_1" + }, + "link": null + }, + { + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [] + } + ], + "properties": { + "proxyWidgets": [ + [ + "94", + "value" + ], + [ + "96", + "value" + ], + [ + "71", + "width" + ], + [ + "71", + "height" + ], + [ + "70", + "seed" + ], + [ + "66", + "unet_name" + ], + [ + "62", + "clip_name" + ], + [ + "98", + "clip_name" + ], + [ + "63", + "vae_name" + ] + ], + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": { + "value": true, + "value_1": true + }, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [], + "title": "Text to Image (Ernie Image Turbo)" + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "2a4f0815-c4d2-4e8b-9bdf-991a8403889d", + "version": 1, + "state": { + "lastGroupId": 7, + "lastNodeId": 103, + "lastLinkId": 134, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "Text to Image (Ernie Image Turbo)", + "inputNode": { + "id": -10, + "bounding": [ + -1350, + 370, + 163.50390625, + 220 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + 1110, + 260, + 120, + 60 + ] + }, + "inputs": [ + { + "id": "74a4609c-67df-4ae9-ab96-9ff4e3a1c3b1", + "name": "value", + "type": "STRING", + "linkIds": [ + 128 + ], + "label": "prompt", + "pos": [ + -1206.49609375, + 390 + ] + }, + { + "id": "996f1854-7ae3-450e-821c-a9b5b7c310f9", + "name": "value_1", + "type": "BOOLEAN", + "linkIds": [ + 127 + ], + "label": "prompt_enhancement", + "pos": [ + -1206.49609375, + 410 + ] + }, + { + "id": "71e9c6e8-4285-4543-b1d3-81520088f6a4", + "name": "width", + "type": "INT", + "linkIds": [ + 104, + 129 + ], + "pos": [ + -1206.49609375, + 430 + ] + }, + { + "id": "bdb6cd97-67d9-440c-8c4c-9b7a7540edd0", + "name": "height", + "type": "INT", + "linkIds": [ + 105, + 130 + ], + "pos": [ + -1206.49609375, + 450 + ] + }, + { + "id": "18abb56c-30bf-4de5-83c1-c12376e8d14e", + "name": "seed", + "type": "INT", + "linkIds": [ + 108 + ], + "pos": [ + -1206.49609375, + 470 + ] + }, + { + "id": "e5cd06f9-64ed-4778-97ba-b165f7a79c4e", + "name": "unet_name", + "type": "COMBO", + "linkIds": [ + 109 + ], + "pos": [ + -1206.49609375, + 490 + ] + }, + { + "id": "06480e4c-4043-489b-ae68-1cf2b4246260", + "name": "clip_name", + "type": "COMBO", + "linkIds": [ + 110 + ], + "pos": [ + -1206.49609375, + 510 + ] + }, + { + "id": "8d65d01b-16b2-420d-8b7b-42077c2e4976", + "name": "clip_name_1", + "type": "COMBO", + "linkIds": [ + 132 + ], + "label": "prompt_enhancer", + "pos": [ + -1206.49609375, + 530 + ] + }, + { + "id": "697f2fdb-0fd9-4008-a895-0f9ce9e8fd88", + "name": "vae_name", + "type": "COMBO", + "linkIds": [ + 133 + ], + "pos": [ + -1206.49609375, + 550 + ] + } + ], + "outputs": [ + { + "id": "21d5fbe0-9f91-4d93-8ea8-5bbf2cd5b698", + "name": "IMAGE", + "type": "IMAGE", + "linkIds": [ + 84 + ], + "localized_name": "IMAGE", + "pos": [ + 1130, + 280 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 71, + "type": "EmptyFlux2LatentImage", + "pos": [ + -470, + 1050 + ], + "size": [ + 270, + 170 + ], + "flags": {}, + "order": 6, + "mode": 0, + "inputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": 104 + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": 105 + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "widget": { + "name": "batch_size" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "links": [ + 80 + ] + } + ], + "properties": { + "Node name for S&R": "EmptyFlux2LatentImage", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + 1024, + 1024, + 1 + ] + }, + { + "id": 66, + "type": "UNETLoader", + "pos": [ + -470, + 320 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "unet_name", + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": 109 + }, + { + "localized_name": "weight_dtype", + "name": "weight_dtype", + "type": "COMBO", + "widget": { + "name": "weight_dtype" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 85 + ] + } + ], + "properties": { + "Node name for S&R": "UNETLoader", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "models": [ + { + "name": "ernie-image-turbo.safetensors", + "url": "https://huggingface.co/Comfy-Org/ERNIE-Image/resolve/main/diffusion_models/ernie-image-turbo.safetensors", + "directory": "diffusion_models" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "ernie-image-turbo.safetensors", + "default" + ] + }, + { + "id": 65, + "type": "VAEDecode", + "pos": [ + 710, + 280 + ], + "size": [ + 230, + 100 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "samples", + "name": "samples", + "type": "LATENT", + "link": 73 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 74 + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "slot_index": 0, + "links": [ + 84 + ] + } + ], + "properties": { + "Node name for S&R": "VAEDecode", + "cnr_id": "comfy-core", + "ver": "0.3.64", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + } + }, + { + "id": 70, + "type": "KSampler", + "pos": [ + 350, + 280 + ], + "size": [ + 320, + 350 + ], + "flags": {}, + "order": 5, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 85 + }, + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "link": 76 + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "link": 113 + }, + { + "localized_name": "latent_image", + "name": "latent_image", + "type": "LATENT", + "link": 80 + }, + { + "localized_name": "seed", + "name": "seed", + "type": "INT", + "widget": { + "name": "seed" + }, + "link": 108 + }, + { + "localized_name": "steps", + "name": "steps", + "type": "INT", + "widget": { + "name": "steps" + }, + "link": null + }, + { + "localized_name": "cfg", + "name": "cfg", + "type": "FLOAT", + "widget": { + "name": "cfg" + }, + "link": null + }, + { + "localized_name": "sampler_name", + "name": "sampler_name", + "type": "COMBO", + "widget": { + "name": "sampler_name" + }, + "link": null + }, + { + "localized_name": "scheduler", + "name": "scheduler", + "type": "COMBO", + "widget": { + "name": "scheduler" + }, + "link": null + }, + { + "localized_name": "denoise", + "name": "denoise", + "type": "FLOAT", + "widget": { + "name": "denoise" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "slot_index": 0, + "links": [ + 73 + ] + } + ], + "properties": { + "Node name for S&R": "KSampler", + "cnr_id": "comfy-core", + "ver": "0.3.64", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + 423299999918804, + "randomize", + 8, + 1, + "euler", + "simple", + 1 + ] + }, + { + "id": 67, + "type": "CLIPTextEncode", + "pos": [ + -140, + 320 + ], + "size": [ + 410, + 370 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 79 + }, + { + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": 131 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 76, + 112 + ] + } + ], + "properties": { + "Node name for S&R": "CLIPTextEncode", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "" + ], + "color": "#232", + "bgcolor": "#353" + }, + { + "id": 62, + "type": "CLIPLoader", + "pos": [ + -470, + 530 + ], + "size": [ + 270, + 150 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "localized_name": "clip_name", + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": 110 + }, + { + "localized_name": "type", + "name": "type", + "type": "COMBO", + "widget": { + "name": "type" + }, + "link": null + }, + { + "localized_name": "device", + "name": "device", + "shape": 7, + "type": "COMBO", + "widget": { + "name": "device" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [ + 79 + ] + } + ], + "properties": { + "Node name for S&R": "CLIPLoader", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "models": [ + { + "name": "ministral-3-3b.safetensors", + "url": "https://huggingface.co/Comfy-Org/ERNIE-Image/resolve/main/text_encoders/ministral-3-3b.safetensors", + "directory": "text_encoders" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "ministral-3-3b.safetensors", + "flux2", + "default" + ] + }, + { + "id": 63, + "type": "VAELoader", + "pos": [ + -470, + 780 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "vae_name", + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": 133 + } + ], + "outputs": [ + { + "localized_name": "VAE", + "name": "VAE", + "type": "VAE", + "links": [ + 74 + ] + } + ], + "properties": { + "Node name for S&R": "VAELoader", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "models": [ + { + "name": "flux2-vae.safetensors", + "url": "https://huggingface.co/Comfy-Org/ERNIE-Image/resolve/main/vae/flux2-vae.safetensors", + "directory": "vae" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "flux2-vae.safetensors" + ] + }, + { + "id": 91, + "type": "ConditioningZeroOut", + "pos": [ + 30, + 760 + ], + "size": [ + 230, + 80 + ], + "flags": {}, + "order": 7, + "mode": 0, + "inputs": [ + { + "localized_name": "conditioning", + "name": "conditioning", + "type": "CONDITIONING", + "link": 112 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 113 + ] + } + ], + "properties": { + "Node name for S&R": "ConditioningZeroOut", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + } + }, + { + "id": 93, + "type": "StringReplace", + "pos": [ + -500, + -650 + ], + "size": [ + 430, + 450 + ], + "flags": {}, + "order": 8, + "mode": 0, + "inputs": [ + { + "localized_name": "string", + "name": "string", + "type": "STRING", + "widget": { + "name": "string" + }, + "link": null + }, + { + "localized_name": "find", + "name": "find", + "type": "STRING", + "widget": { + "name": "find" + }, + "link": null + }, + { + "localized_name": "replace", + "name": "replace", + "type": "STRING", + "widget": { + "name": "replace" + }, + "link": 115 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [ + 121 + ] + } + ], + "properties": { + "Node name for S&R": "StringReplace", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "[SYSTEM_PROMPT]你是一个专业的文生图 Prompt 增强助手。你将收到用户的简短图片描述及目标生成分辨率，请据此扩写为一段内容丰富、细节充分的视觉描述，以帮助文生图模型生成高质量的图片。仅输出增强后的描述，不要包含任何解释或前缀。[/SYSTEM_PROMPT][INST]{\"prompt\": \"{prompt}\", \"width\": {width}, \"height\": {height}}[/INST]", + "{prompt}", + "" + ] + }, + { + "id": 94, + "type": "PrimitiveStringMultiline", + "pos": [ + -950, + -660 + ], + "size": [ + 400, + 200 + ], + "flags": {}, + "order": 9, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "STRING", + "widget": { + "name": "value" + }, + "link": 128 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [ + 115, + 118 + ] + } + ], + "title": "String (Multiline - Prompt)", + "properties": { + "Node name for S&R": "PrimitiveStringMultiline", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "" + ] + }, + { + "id": 95, + "type": "TextGenerate", + "pos": [ + 530, + -660 + ], + "size": [ + 400, + 380 + ], + "flags": {}, + "order": 10, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 116 + }, + { + "localized_name": "image", + "name": "image", + "shape": 7, + "type": "IMAGE", + "link": null + }, + { + "localized_name": "prompt", + "name": "prompt", + "type": "STRING", + "widget": { + "name": "prompt" + }, + "link": 117 + }, + { + "localized_name": "max_length", + "name": "max_length", + "type": "INT", + "widget": { + "name": "max_length" + }, + "link": null + }, + { + "localized_name": "sampling_mode", + "name": "sampling_mode", + "type": "COMFY_DYNAMICCOMBO_V3", + "widget": { + "name": "sampling_mode" + }, + "link": null + }, + { + "localized_name": "temperature", + "name": "sampling_mode.temperature", + "type": "FLOAT", + "widget": { + "name": "sampling_mode.temperature" + }, + "link": null + }, + { + "localized_name": "top_k", + "name": "sampling_mode.top_k", + "type": "INT", + "widget": { + "name": "sampling_mode.top_k" + }, + "link": null + }, + { + "localized_name": "top_p", + "name": "sampling_mode.top_p", + "type": "FLOAT", + "widget": { + "name": "sampling_mode.top_p" + }, + "link": null + }, + { + "localized_name": "min_p", + "name": "sampling_mode.min_p", + "type": "FLOAT", + "widget": { + "name": "sampling_mode.min_p" + }, + "link": null + }, + { + "localized_name": "repetition_penalty", + "name": "sampling_mode.repetition_penalty", + "type": "FLOAT", + "widget": { + "name": "sampling_mode.repetition_penalty" + }, + "link": null + }, + { + "localized_name": "seed", + "name": "sampling_mode.seed", + "type": "INT", + "widget": { + "name": "sampling_mode.seed" + }, + "link": null + }, + { + "localized_name": "sampling_mode.presence_penalty", + "name": "sampling_mode.presence_penalty", + "shape": 7, + "type": "FLOAT", + "widget": { + "name": "sampling_mode.presence_penalty" + }, + "link": null + }, + { + "localized_name": "thinking", + "name": "thinking", + "shape": 7, + "type": "BOOLEAN", + "widget": { + "name": "thinking" + }, + "link": null + }, + { + "localized_name": "use_default_template", + "name": "use_default_template", + "shape": 7, + "type": "BOOLEAN", + "widget": { + "name": "use_default_template" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "generated_text", + "name": "generated_text", + "type": "STRING", + "links": [ + 119 + ] + } + ], + "properties": { + "Node name for S&R": "TextGenerate", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "", + 2048, + "on", + 0.6, + 64, + 0.8, + 0.05, + 1.05, + 0, + 0, + false, + true + ] + }, + { + "id": 96, + "type": "PrimitiveBoolean", + "pos": [ + -490, + 60 + ], + "size": [ + 270, + 100 + ], + "flags": {}, + "order": 11, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "BOOLEAN", + "widget": { + "name": "value" + }, + "link": 127 + } + ], + "outputs": [ + { + "localized_name": "BOOLEAN", + "name": "BOOLEAN", + "type": "BOOLEAN", + "links": [ + 120 + ] + } + ], + "title": "Enable prompt enhancement?", + "properties": { + "Node name for S&R": "PrimitiveBoolean", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + true + ] + }, + { + "id": 97, + "type": "ComfySwitchNode", + "pos": [ + 550, + -10 + ], + "size": [ + 270, + 130 + ], + "flags": {}, + "order": 12, + "mode": 0, + "inputs": [ + { + "localized_name": "on_false", + "name": "on_false", + "type": "*", + "link": 118 + }, + { + "localized_name": "on_true", + "name": "on_true", + "type": "*", + "link": 119 + }, + { + "localized_name": "switch", + "name": "switch", + "type": "BOOLEAN", + "widget": { + "name": "switch" + }, + "link": 120 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "*", + "links": [ + 131, + 134 + ] + } + ], + "properties": { + "Node name for S&R": "ComfySwitchNode", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + false + ] + }, + { + "id": 98, + "type": "CLIPLoader", + "pos": [ + -490, + -150 + ], + "size": [ + 510, + 150 + ], + "flags": {}, + "order": 13, + "mode": 0, + "inputs": [ + { + "localized_name": "clip_name", + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": 132 + }, + { + "localized_name": "type", + "name": "type", + "type": "COMBO", + "widget": { + "name": "type" + }, + "link": null + }, + { + "localized_name": "device", + "name": "device", + "shape": 7, + "type": "COMBO", + "widget": { + "name": "device" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [ + 116 + ] + } + ], + "title": "Load CLIP (PE)", + "properties": { + "Node name for S&R": "CLIPLoader", + "cnr_id": "comfy-core", + "ver": "0.19.0", + "models": [ + { + "name": "ernie-image-prompt-enhancer.safetensors", + "url": "https://huggingface.co/Comfy-Org/ERNIE-Image/resolve/main/text_encoders/ernie-image-prompt-enhancer.safetensors", + "directory": "text_encoders" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "ernie-image-prompt-enhancer.safetensors", + "flux2", + "default" + ] + }, + { + "id": 99, + "type": "PreviewAny", + "pos": [ + -950, + -410 + ], + "size": [ + 400, + 180 + ], + "flags": {}, + "order": 14, + "mode": 0, + "inputs": [ + { + "localized_name": "source", + "name": "source", + "type": "*", + "link": 129 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [ + 122 + ] + } + ], + "title": "Preview as Text (Int to String)", + "properties": { + "Node name for S&R": "PreviewAny", + "cnr_id": "comfy-core", + "ver": "0.19.0", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + null, + null, + null + ] + }, + { + "id": 100, + "type": "PreviewAny", + "pos": [ + -950, + -190 + ], + "size": [ + 400, + 180 + ], + "flags": {}, + "order": 15, + "mode": 0, + "inputs": [ + { + "localized_name": "source", + "name": "source", + "type": "*", + "link": 130 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [ + 124 + ] + } + ], + "title": "Preview as Text (Int to String)", + "properties": { + "Node name for S&R": "PreviewAny", + "cnr_id": "comfy-core", + "ver": "0.19.0", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + null, + null, + null + ] + }, + { + "id": 101, + "type": "StringReplace", + "pos": [ + -30, + -650 + ], + "size": [ + 230, + 450 + ], + "flags": {}, + "order": 16, + "mode": 0, + "inputs": [ + { + "localized_name": "string", + "name": "string", + "type": "STRING", + "widget": { + "name": "string" + }, + "link": 121 + }, + { + "localized_name": "find", + "name": "find", + "type": "STRING", + "widget": { + "name": "find" + }, + "link": null + }, + { + "localized_name": "replace", + "name": "replace", + "type": "STRING", + "widget": { + "name": "replace" + }, + "link": 122 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [ + 123 + ] + } + ], + "properties": { + "Node name for S&R": "StringReplace", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "", + "{width}", + "" + ] + }, + { + "id": 102, + "type": "StringReplace", + "pos": [ + 220, + -650 + ], + "size": [ + 250, + 450 + ], + "flags": {}, + "order": 17, + "mode": 0, + "inputs": [ + { + "localized_name": "string", + "name": "string", + "type": "STRING", + "widget": { + "name": "string" + }, + "link": 123 + }, + { + "localized_name": "find", + "name": "find", + "type": "STRING", + "widget": { + "name": "find" + }, + "link": null + }, + { + "localized_name": "replace", + "name": "replace", + "type": "STRING", + "widget": { + "name": "replace" + }, + "link": 124 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [ + 117 + ] + } + ], + "properties": { + "Node name for S&R": "StringReplace", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "", + "{height}", + "" + ] + }, + { + "id": 103, + "type": "PreviewAny", + "pos": [ + 970, + -660 + ], + "size": [ + 570, + 790 + ], + "flags": {}, + "order": 18, + "mode": 0, + "inputs": [ + { + "localized_name": "source", + "name": "source", + "type": "*", + "link": 134 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [] + } + ], + "title": "Preview as Text (Int to String)", + "properties": { + "Node name for S&R": "PreviewAny", + "cnr_id": "comfy-core", + "ver": "0.19.0", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + null, + null, + null + ] + } + ], + "groups": [ + { + "id": 6, + "title": "Text to Image", + "bounding": [ + -510, + 200, + 1450, + 1060 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 2, + "title": "Image Size", + "bounding": [ + -490, + 950, + 300, + 290 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 3, + "title": "Prompt", + "bounding": [ + -160, + 250, + 470, + 670 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 4, + "title": "Model", + "bounding": [ + -490, + 250, + 300, + 670 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 7, + "title": "Prompt Enhancement", + "bounding": [ + -510, + -720, + 1450, + 890 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + } + ], + "links": [ + { + "id": 73, + "origin_id": 70, + "origin_slot": 0, + "target_id": 65, + "target_slot": 0, + "type": "LATENT" + }, + { + "id": 74, + "origin_id": 63, + "origin_slot": 0, + "target_id": 65, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 85, + "origin_id": 66, + "origin_slot": 0, + "target_id": 70, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 76, + "origin_id": 67, + "origin_slot": 0, + "target_id": 70, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 80, + "origin_id": 71, + "origin_slot": 0, + "target_id": 70, + "target_slot": 3, + "type": "LATENT" + }, + { + "id": 79, + "origin_id": 62, + "origin_slot": 0, + "target_id": 67, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 84, + "origin_id": 65, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 104, + "origin_id": -10, + "origin_slot": 2, + "target_id": 71, + "target_slot": 0, + "type": "INT" + }, + { + "id": 105, + "origin_id": -10, + "origin_slot": 3, + "target_id": 71, + "target_slot": 1, + "type": "INT" + }, + { + "id": 108, + "origin_id": -10, + "origin_slot": 4, + "target_id": 70, + "target_slot": 4, + "type": "INT" + }, + { + "id": 109, + "origin_id": -10, + "origin_slot": 5, + "target_id": 66, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 110, + "origin_id": -10, + "origin_slot": 6, + "target_id": 62, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 112, + "origin_id": 67, + "origin_slot": 0, + "target_id": 91, + "target_slot": 0, + "type": "CONDITIONING" + }, + { + "id": 113, + "origin_id": 91, + "origin_slot": 0, + "target_id": 70, + "target_slot": 2, + "type": "CONDITIONING" + }, + { + "id": 115, + "origin_id": 94, + "origin_slot": 0, + "target_id": 93, + "target_slot": 2, + "type": "STRING" + }, + { + "id": 116, + "origin_id": 98, + "origin_slot": 0, + "target_id": 95, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 117, + "origin_id": 102, + "origin_slot": 0, + "target_id": 95, + "target_slot": 2, + "type": "STRING" + }, + { + "id": 118, + "origin_id": 94, + "origin_slot": 0, + "target_id": 97, + "target_slot": 0, + "type": "STRING" + }, + { + "id": 119, + "origin_id": 95, + "origin_slot": 0, + "target_id": 97, + "target_slot": 1, + "type": "STRING" + }, + { + "id": 120, + "origin_id": 96, + "origin_slot": 0, + "target_id": 97, + "target_slot": 2, + "type": "BOOLEAN" + }, + { + "id": 121, + "origin_id": 93, + "origin_slot": 0, + "target_id": 101, + "target_slot": 0, + "type": "STRING" + }, + { + "id": 122, + "origin_id": 99, + "origin_slot": 0, + "target_id": 101, + "target_slot": 2, + "type": "STRING" + }, + { + "id": 123, + "origin_id": 101, + "origin_slot": 0, + "target_id": 102, + "target_slot": 0, + "type": "STRING" + }, + { + "id": 124, + "origin_id": 100, + "origin_slot": 0, + "target_id": 102, + "target_slot": 2, + "type": "STRING" + }, + { + "id": 127, + "origin_id": -10, + "origin_slot": 1, + "target_id": 96, + "target_slot": 0, + "type": "BOOLEAN" + }, + { + "id": 128, + "origin_id": -10, + "origin_slot": 0, + "target_id": 94, + "target_slot": 0, + "type": "STRING" + }, + { + "id": 129, + "origin_id": -10, + "origin_slot": 2, + "target_id": 99, + "target_slot": 0, + "type": "*" + }, + { + "id": 130, + "origin_id": -10, + "origin_slot": 3, + "target_id": 100, + "target_slot": 0, + "type": "*" + }, + { + "id": 131, + "origin_id": 97, + "origin_slot": 0, + "target_id": 67, + "target_slot": 1, + "type": "STRING" + }, + { + "id": 132, + "origin_id": -10, + "origin_slot": 7, + "target_id": 98, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 133, + "origin_id": -10, + "origin_slot": 8, + "target_id": 63, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 134, + "origin_id": 97, + "origin_slot": 0, + "target_id": 103, + "target_slot": 0, + "type": "STRING" + } + ], + "extra": {}, + "category": "Image generation and editing/Text to image", + "description": "Faster ERNIE Image Turbo variant (~8B DiT, distilled for fewer sampling steps): same strengths in Chinese/English on-image text and layout-heavy graphics as the base ERNIE Image lineup, with bundled encoders and VAE." + } + ] + }, + "extra": { + "ue_links": [] + } +} diff --git a/blueprints/Text to Image (Ernie Image).json b/blueprints/Text to Image (Ernie Image).json new file mode 100644 index 000000000..2bab20d69 --- /dev/null +++ b/blueprints/Text to Image (Ernie Image).json @@ -0,0 +1,2190 @@ +{ + "revision": 0, + "last_node_id": 88, + "last_link_id": 0, + "nodes": [ + { + "id": 88, + "type": "03921aea-a70e-44b4-bc77-f6bda10f2120", + "pos": [ + -120, + 240 + ], + "size": [ + 400, + 540 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "label": "prompt", + "name": "value", + "type": "STRING", + "widget": { + "name": "value" + }, + "link": null + }, + { + "label": "prompt_enhancement", + "name": "value_1", + "type": "BOOLEAN", + "widget": { + "name": "value_1" + }, + "link": null + }, + { + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": null + }, + { + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": null + }, + { + "name": "steps", + "type": "INT", + "widget": { + "name": "steps" + }, + "link": null + }, + { + "name": "cfg", + "type": "FLOAT", + "widget": { + "name": "cfg" + }, + "link": null + }, + { + "name": "seed", + "type": "INT", + "widget": { + "name": "seed" + }, + "link": null + }, + { + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": null + }, + { + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": null + }, + { + "label": "prompt_enhancer", + "name": "clip_name_1", + "type": "COMBO", + "widget": { + "name": "clip_name_1" + }, + "link": null + }, + { + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [] + } + ], + "properties": { + "proxyWidgets": [ + [ + "78", + "value" + ], + [ + "76", + "value" + ], + [ + "71", + "width" + ], + [ + "71", + "height" + ], + [ + "70", + "steps" + ], + [ + "70", + "cfg" + ], + [ + "70", + "seed" + ], + [ + "66", + "unet_name" + ], + [ + "62", + "clip_name" + ], + [ + "91", + "clip_name" + ], + [ + "63", + "vae_name" + ] + ], + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": { + "value": true, + "value_1": true + }, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [], + "title": "Text to Image (Ernie Image)" + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "03921aea-a70e-44b4-bc77-f6bda10f2120", + "version": 1, + "state": { + "lastGroupId": 6, + "lastNodeId": 99, + "lastLinkId": 124, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "Text to Image (Ernie Image)", + "inputNode": { + "id": -10, + "bounding": [ + -1350, + 370, + 163.50390625, + 260 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + 1110, + 260, + 120, + 60 + ] + }, + "inputs": [ + { + "id": "504de359-52a4-49aa-b6be-23c1cdb0cbde", + "name": "value", + "type": "STRING", + "linkIds": [ + 102 + ], + "label": "prompt", + "pos": [ + -1206.49609375, + 390 + ] + }, + { + "id": "29f699c6-9263-41f6-b37d-69b9fc3913dd", + "name": "value_1", + "type": "BOOLEAN", + "linkIds": [ + 103 + ], + "label": "prompt_enhancement", + "pos": [ + -1206.49609375, + 410 + ] + }, + { + "id": "968e6213-d1e9-4268-8f47-1d6b9a39a43e", + "name": "width", + "type": "INT", + "linkIds": [ + 104, + 113 + ], + "pos": [ + -1206.49609375, + 430 + ] + }, + { + "id": "181c49ef-740d-4385-aa11-79718951ccb9", + "name": "height", + "type": "INT", + "linkIds": [ + 105, + 114 + ], + "pos": [ + -1206.49609375, + 450 + ] + }, + { + "id": "1e85f808-66a1-41df-be52-334142b35419", + "name": "steps", + "type": "INT", + "linkIds": [ + 106 + ], + "pos": [ + -1206.49609375, + 470 + ] + }, + { + "id": "2806addf-a252-4aa3-a5b7-397ab36dccec", + "name": "cfg", + "type": "FLOAT", + "linkIds": [ + 107 + ], + "pos": [ + -1206.49609375, + 490 + ] + }, + { + "id": "5d036a66-5dc0-4d7c-b9a9-349e454738aa", + "name": "seed", + "type": "INT", + "linkIds": [ + 108 + ], + "pos": [ + -1206.49609375, + 510 + ] + }, + { + "id": "360f9a40-aac5-4e9c-bc98-9d55a4a58be2", + "name": "unet_name", + "type": "COMBO", + "linkIds": [ + 109 + ], + "pos": [ + -1206.49609375, + 530 + ] + }, + { + "id": "886301c7-6e88-4cec-96fa-8ae20e8340c5", + "name": "clip_name", + "type": "COMBO", + "linkIds": [ + 110 + ], + "pos": [ + -1206.49609375, + 550 + ] + }, + { + "id": "1d73a545-6d01-462f-bc61-966d4b918ff2", + "name": "clip_name_1", + "type": "COMBO", + "linkIds": [ + 120 + ], + "label": "prompt_enhancer", + "pos": [ + -1206.49609375, + 570 + ] + }, + { + "id": "8c61dc8c-e260-4b36-b73e-d36f90a0bbe3", + "name": "vae_name", + "type": "COMBO", + "linkIds": [ + 121 + ], + "pos": [ + -1206.49609375, + 590 + ] + } + ], + "outputs": [ + { + "id": "f4cb34c8-4090-4281-b428-7338a339d274", + "name": "IMAGE", + "type": "IMAGE", + "linkIds": [ + 84 + ], + "localized_name": "IMAGE", + "pos": [ + 1130, + 280 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 71, + "type": "EmptyFlux2LatentImage", + "pos": [ + -460, + 1040 + ], + "size": [ + 270, + 170 + ], + "flags": {}, + "order": 6, + "mode": 0, + "inputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": 104 + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": 105 + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "widget": { + "name": "batch_size" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "links": [ + 80 + ] + } + ], + "properties": { + "Node name for S&R": "EmptyFlux2LatentImage", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + 1024, + 1024, + 1 + ] + }, + { + "id": 66, + "type": "UNETLoader", + "pos": [ + -470, + 320 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "unet_name", + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": 109 + }, + { + "localized_name": "weight_dtype", + "name": "weight_dtype", + "type": "COMBO", + "widget": { + "name": "weight_dtype" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 85 + ] + } + ], + "properties": { + "Node name for S&R": "UNETLoader", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "models": [ + { + "name": "ernie-image.safetensors", + "url": "https://huggingface.co/Comfy-Org/ERNIE-Image/resolve/main/diffusion_models/ernie-image.safetensors", + "directory": "diffusion_models" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "ernie-image.safetensors", + "default" + ] + }, + { + "id": 65, + "type": "VAEDecode", + "pos": [ + 710, + 280 + ], + "size": [ + 230, + 100 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "samples", + "name": "samples", + "type": "LATENT", + "link": 73 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 74 + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "slot_index": 0, + "links": [ + 84 + ] + } + ], + "properties": { + "Node name for S&R": "VAEDecode", + "cnr_id": "comfy-core", + "ver": "0.3.64", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + } + }, + { + "id": 70, + "type": "KSampler", + "pos": [ + 350, + 280 + ], + "size": [ + 320, + 350 + ], + "flags": {}, + "order": 5, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 85 + }, + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "link": 76 + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "link": 83 + }, + { + "localized_name": "latent_image", + "name": "latent_image", + "type": "LATENT", + "link": 80 + }, + { + "localized_name": "seed", + "name": "seed", + "type": "INT", + "widget": { + "name": "seed" + }, + "link": 108 + }, + { + "localized_name": "steps", + "name": "steps", + "type": "INT", + "widget": { + "name": "steps" + }, + "link": 106 + }, + { + "localized_name": "cfg", + "name": "cfg", + "type": "FLOAT", + "widget": { + "name": "cfg" + }, + "link": 107 + }, + { + "localized_name": "sampler_name", + "name": "sampler_name", + "type": "COMBO", + "widget": { + "name": "sampler_name" + }, + "link": null + }, + { + "localized_name": "scheduler", + "name": "scheduler", + "type": "COMBO", + "widget": { + "name": "scheduler" + }, + "link": null + }, + { + "localized_name": "denoise", + "name": "denoise", + "type": "FLOAT", + "widget": { + "name": "denoise" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "slot_index": 0, + "links": [ + 73 + ] + } + ], + "properties": { + "Node name for S&R": "KSampler", + "cnr_id": "comfy-core", + "ver": "0.3.64", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + 182596410725960, + "randomize", + 20, + 4, + "euler", + "simple", + 1 + ] + }, + { + "id": 67, + "type": "CLIPTextEncode", + "pos": [ + -140, + 320 + ], + "size": [ + 410, + 370 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 79 + }, + { + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": 100 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 76 + ] + } + ], + "properties": { + "Node name for S&R": "CLIPTextEncode", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "" + ], + "color": "#232", + "bgcolor": "#353" + }, + { + "id": 72, + "type": "CLIPTextEncode", + "pos": [ + -130, + 770 + ], + "size": [ + 390, + 140 + ], + "flags": {}, + "order": 7, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 82 + }, + { + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 83 + ] + } + ], + "properties": { + "Node name for S&R": "CLIPTextEncode", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "" + ], + "color": "#223", + "bgcolor": "#335" + }, + { + "id": 83, + "type": "StringReplace", + "pos": [ + -500, + -640 + ], + "size": [ + 430, + 450 + ], + "flags": {}, + "order": 12, + "mode": 0, + "inputs": [ + { + "localized_name": "string", + "name": "string", + "type": "STRING", + "widget": { + "name": "string" + }, + "link": null + }, + { + "localized_name": "find", + "name": "find", + "type": "STRING", + "widget": { + "name": "find" + }, + "link": null + }, + { + "localized_name": "replace", + "name": "replace", + "type": "STRING", + "widget": { + "name": "replace" + }, + "link": 92 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [ + 115 + ] + } + ], + "properties": { + "Node name for S&R": "StringReplace", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "[SYSTEM_PROMPT]你是一个专业的文生图 Prompt 增强助手。你将收到用户的简短图片描述及目标生成分辨率，请据此扩写为一段内容丰富、细节充分的视觉描述，以帮助文生图模型生成高质量的图片。仅输出增强后的描述，不要包含任何解释或前缀。[/SYSTEM_PROMPT][INST]{\"prompt\": \"{prompt}\", \"width\": {width}, \"height\": {height}}[/INST]", + "{prompt}", + "" + ] + }, + { + "id": 78, + "type": "PrimitiveStringMultiline", + "pos": [ + -950, + -650 + ], + "size": [ + 400, + 200 + ], + "flags": {}, + "order": 11, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "STRING", + "widget": { + "name": "value" + }, + "link": 102 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [ + 87, + 92 + ] + } + ], + "title": "String (Multiline - Prompt)", + "properties": { + "Node name for S&R": "PrimitiveStringMultiline", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "" + ] + }, + { + "id": 74, + "type": "TextGenerate", + "pos": [ + 530, + -650 + ], + "size": [ + 400, + 380 + ], + "flags": {}, + "order": 8, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 112 + }, + { + "localized_name": "image", + "name": "image", + "shape": 7, + "type": "IMAGE", + "link": null + }, + { + "localized_name": "prompt", + "name": "prompt", + "type": "STRING", + "widget": { + "name": "prompt" + }, + "link": 119 + }, + { + "localized_name": "max_length", + "name": "max_length", + "type": "INT", + "widget": { + "name": "max_length" + }, + "link": null + }, + { + "localized_name": "sampling_mode", + "name": "sampling_mode", + "type": "COMFY_DYNAMICCOMBO_V3", + "widget": { + "name": "sampling_mode" + }, + "link": null + }, + { + "localized_name": "temperature", + "name": "sampling_mode.temperature", + "type": "FLOAT", + "widget": { + "name": "sampling_mode.temperature" + }, + "link": null + }, + { + "localized_name": "top_k", + "name": "sampling_mode.top_k", + "type": "INT", + "widget": { + "name": "sampling_mode.top_k" + }, + "link": null + }, + { + "localized_name": "top_p", + "name": "sampling_mode.top_p", + "type": "FLOAT", + "widget": { + "name": "sampling_mode.top_p" + }, + "link": null + }, + { + "localized_name": "min_p", + "name": "sampling_mode.min_p", + "type": "FLOAT", + "widget": { + "name": "sampling_mode.min_p" + }, + "link": null + }, + { + "localized_name": "repetition_penalty", + "name": "sampling_mode.repetition_penalty", + "type": "FLOAT", + "widget": { + "name": "sampling_mode.repetition_penalty" + }, + "link": null + }, + { + "localized_name": "seed", + "name": "sampling_mode.seed", + "type": "INT", + "widget": { + "name": "sampling_mode.seed" + }, + "link": null + }, + { + "localized_name": "sampling_mode.presence_penalty", + "name": "sampling_mode.presence_penalty", + "shape": 7, + "type": "FLOAT", + "widget": { + "name": "sampling_mode.presence_penalty" + }, + "link": null + }, + { + "localized_name": "thinking", + "name": "thinking", + "shape": 7, + "type": "BOOLEAN", + "widget": { + "name": "thinking" + }, + "link": null + }, + { + "localized_name": "use_default_template", + "name": "use_default_template", + "shape": 7, + "type": "BOOLEAN", + "widget": { + "name": "use_default_template" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "generated_text", + "name": "generated_text", + "type": "STRING", + "links": [ + 89 + ] + } + ], + "properties": { + "Node name for S&R": "TextGenerate", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "", + 2048, + "on", + 0.6, + 64, + 0.8, + 0.05, + 1.05, + 0, + 0, + false, + true + ] + }, + { + "id": 76, + "type": "PrimitiveBoolean", + "pos": [ + -500, + 60 + ], + "size": [ + 270, + 100 + ], + "flags": {}, + "order": 10, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "BOOLEAN", + "widget": { + "name": "value" + }, + "link": 103 + } + ], + "outputs": [ + { + "localized_name": "BOOLEAN", + "name": "BOOLEAN", + "type": "BOOLEAN", + "links": [ + 88 + ] + } + ], + "title": "Enable prompt enhancement?", + "properties": { + "Node name for S&R": "PrimitiveBoolean", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + true + ] + }, + { + "id": 75, + "type": "ComfySwitchNode", + "pos": [ + 530, + 20 + ], + "size": [ + 270, + 130 + ], + "flags": {}, + "order": 9, + "mode": 0, + "inputs": [ + { + "localized_name": "on_false", + "name": "on_false", + "type": "*", + "link": 87 + }, + { + "localized_name": "on_true", + "name": "on_true", + "type": "*", + "link": 89 + }, + { + "localized_name": "switch", + "name": "switch", + "type": "BOOLEAN", + "widget": { + "name": "switch" + }, + "link": 88 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "*", + "links": [ + 100, + 124 + ] + } + ], + "properties": { + "Node name for S&R": "ComfySwitchNode", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + false + ] + }, + { + "id": 62, + "type": "CLIPLoader", + "pos": [ + -460, + 520 + ], + "size": [ + 270, + 150 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "localized_name": "clip_name", + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": 110 + }, + { + "localized_name": "type", + "name": "type", + "type": "COMBO", + "widget": { + "name": "type" + }, + "link": null + }, + { + "localized_name": "device", + "name": "device", + "shape": 7, + "type": "COMBO", + "widget": { + "name": "device" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [ + 79, + 82 + ] + } + ], + "properties": { + "Node name for S&R": "CLIPLoader", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "models": [ + { + "name": "ministral-3-3b.safetensors", + "url": "https://huggingface.co/Comfy-Org/ERNIE-Image/resolve/main/text_encoders/ministral-3-3b.safetensors", + "directory": "text_encoders" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "ministral-3-3b.safetensors", + "flux2", + "default" + ] + }, + { + "id": 63, + "type": "VAELoader", + "pos": [ + -460, + 770 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "vae_name", + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": 121 + } + ], + "outputs": [ + { + "localized_name": "VAE", + "name": "VAE", + "type": "VAE", + "links": [ + 74 + ] + } + ], + "properties": { + "Node name for S&R": "VAELoader", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "models": [ + { + "name": "flux2-vae.safetensors", + "url": "https://huggingface.co/Comfy-Org/ERNIE-Image/resolve/main/vae/flux2-vae.safetensors", + "directory": "vae" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "flux2-vae.safetensors" + ] + }, + { + "id": 91, + "type": "CLIPLoader", + "pos": [ + -500, + -150 + ], + "size": [ + 510, + 150 + ], + "flags": {}, + "order": 13, + "mode": 0, + "inputs": [ + { + "localized_name": "clip_name", + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": 120 + }, + { + "localized_name": "type", + "name": "type", + "type": "COMBO", + "widget": { + "name": "type" + }, + "link": null + }, + { + "localized_name": "device", + "name": "device", + "shape": 7, + "type": "COMBO", + "widget": { + "name": "device" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [ + 112 + ] + } + ], + "title": "Load CLIP (PE)", + "properties": { + "Node name for S&R": "CLIPLoader", + "cnr_id": "comfy-core", + "ver": "0.19.0", + "models": [ + { + "name": "ernie-image-prompt-enhancer.safetensors", + "url": "https://huggingface.co/Comfy-Org/ERNIE-Image/resolve/main/text_encoders/ernie-image-prompt-enhancer.safetensors", + "directory": "text_encoders" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "ernie-image-prompt-enhancer.safetensors", + "flux2", + "default" + ] + }, + { + "id": 92, + "type": "PreviewAny", + "pos": [ + -950, + -400 + ], + "size": [ + 400, + 180 + ], + "flags": {}, + "order": 14, + "mode": 0, + "inputs": [ + { + "localized_name": "source", + "name": "source", + "type": "*", + "link": 113 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [ + 116 + ] + } + ], + "title": "Preview as Text (Int to String)", + "properties": { + "Node name for S&R": "PreviewAny", + "cnr_id": "comfy-core", + "ver": "0.19.0", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + null, + null, + null + ] + }, + { + "id": 93, + "type": "PreviewAny", + "pos": [ + -950, + -180 + ], + "size": [ + 400, + 180 + ], + "flags": {}, + "order": 15, + "mode": 0, + "inputs": [ + { + "localized_name": "source", + "name": "source", + "type": "*", + "link": 114 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [ + 118 + ] + } + ], + "title": "Preview as Text (Int to String)", + "properties": { + "Node name for S&R": "PreviewAny", + "cnr_id": "comfy-core", + "ver": "0.19.0", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + null, + null, + null + ] + }, + { + "id": 94, + "type": "StringReplace", + "pos": [ + -30, + -640 + ], + "size": [ + 230, + 450 + ], + "flags": {}, + "order": 16, + "mode": 0, + "inputs": [ + { + "localized_name": "string", + "name": "string", + "type": "STRING", + "widget": { + "name": "string" + }, + "link": 115 + }, + { + "localized_name": "find", + "name": "find", + "type": "STRING", + "widget": { + "name": "find" + }, + "link": null + }, + { + "localized_name": "replace", + "name": "replace", + "type": "STRING", + "widget": { + "name": "replace" + }, + "link": 116 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [ + 117 + ] + } + ], + "properties": { + "Node name for S&R": "StringReplace", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "", + "{width}", + "" + ] + }, + { + "id": 95, + "type": "StringReplace", + "pos": [ + 220, + -640 + ], + "size": [ + 250, + 450 + ], + "flags": {}, + "order": 17, + "mode": 0, + "inputs": [ + { + "localized_name": "string", + "name": "string", + "type": "STRING", + "widget": { + "name": "string" + }, + "link": 117 + }, + { + "localized_name": "find", + "name": "find", + "type": "STRING", + "widget": { + "name": "find" + }, + "link": null + }, + { + "localized_name": "replace", + "name": "replace", + "type": "STRING", + "widget": { + "name": "replace" + }, + "link": 118 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [ + 119 + ] + } + ], + "properties": { + "Node name for S&R": "StringReplace", + "cnr_id": "comfy-core", + "ver": "0.18.1", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + "", + "{height}", + "" + ] + }, + { + "id": 97, + "type": "PreviewAny", + "pos": [ + 970, + -650 + ], + "size": [ + 570, + 790 + ], + "flags": {}, + "order": 18, + "mode": 0, + "inputs": [ + { + "localized_name": "source", + "name": "source", + "type": "*", + "link": 124 + } + ], + "outputs": [ + { + "localized_name": "STRING", + "name": "STRING", + "type": "STRING", + "links": [] + } + ], + "title": "Preview as Text (Int to String)", + "properties": { + "Node name for S&R": "PreviewAny", + "cnr_id": "comfy-core", + "ver": "0.19.0", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + } + }, + "widgets_values": [ + null, + null, + null + ] + } + ], + "groups": [ + { + "id": 6, + "title": "Text to Image", + "bounding": [ + -510, + 200, + 1450, + 1060 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 2, + "title": "Image Size", + "bounding": [ + -480, + 940, + 310, + 290 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 3, + "title": "Prompt", + "bounding": [ + -160, + 250, + 470, + 670 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 4, + "title": "Model", + "bounding": [ + -490, + 250, + 320, + 670 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 5, + "title": "Prompt Enhancement", + "bounding": [ + -510, + -720, + 1450, + 890 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + } + ], + "links": [ + { + "id": 73, + "origin_id": 70, + "origin_slot": 0, + "target_id": 65, + "target_slot": 0, + "type": "LATENT" + }, + { + "id": 74, + "origin_id": 63, + "origin_slot": 0, + "target_id": 65, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 85, + "origin_id": 66, + "origin_slot": 0, + "target_id": 70, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 76, + "origin_id": 67, + "origin_slot": 0, + "target_id": 70, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 83, + "origin_id": 72, + "origin_slot": 0, + "target_id": 70, + "target_slot": 2, + "type": "CONDITIONING" + }, + { + "id": 80, + "origin_id": 71, + "origin_slot": 0, + "target_id": 70, + "target_slot": 3, + "type": "LATENT" + }, + { + "id": 79, + "origin_id": 62, + "origin_slot": 0, + "target_id": 67, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 100, + "origin_id": 75, + "origin_slot": 0, + "target_id": 67, + "target_slot": 1, + "type": "STRING" + }, + { + "id": 82, + "origin_id": 62, + "origin_slot": 0, + "target_id": 72, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 92, + "origin_id": 78, + "origin_slot": 0, + "target_id": 83, + "target_slot": 2, + "type": "STRING" + }, + { + "id": 87, + "origin_id": 78, + "origin_slot": 0, + "target_id": 75, + "target_slot": 0, + "type": "STRING" + }, + { + "id": 89, + "origin_id": 74, + "origin_slot": 0, + "target_id": 75, + "target_slot": 1, + "type": "STRING" + }, + { + "id": 88, + "origin_id": 76, + "origin_slot": 0, + "target_id": 75, + "target_slot": 2, + "type": "BOOLEAN" + }, + { + "id": 84, + "origin_id": 65, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 102, + "origin_id": -10, + "origin_slot": 0, + "target_id": 78, + "target_slot": 0, + "type": "STRING" + }, + { + "id": 103, + "origin_id": -10, + "origin_slot": 1, + "target_id": 76, + "target_slot": 0, + "type": "BOOLEAN" + }, + { + "id": 104, + "origin_id": -10, + "origin_slot": 2, + "target_id": 71, + "target_slot": 0, + "type": "INT" + }, + { + "id": 105, + "origin_id": -10, + "origin_slot": 3, + "target_id": 71, + "target_slot": 1, + "type": "INT" + }, + { + "id": 106, + "origin_id": -10, + "origin_slot": 4, + "target_id": 70, + "target_slot": 5, + "type": "INT" + }, + { + "id": 107, + "origin_id": -10, + "origin_slot": 5, + "target_id": 70, + "target_slot": 6, + "type": "FLOAT" + }, + { + "id": 108, + "origin_id": -10, + "origin_slot": 6, + "target_id": 70, + "target_slot": 4, + "type": "INT" + }, + { + "id": 109, + "origin_id": -10, + "origin_slot": 7, + "target_id": 66, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 110, + "origin_id": -10, + "origin_slot": 8, + "target_id": 62, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 112, + "origin_id": 91, + "origin_slot": 0, + "target_id": 74, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 113, + "origin_id": -10, + "origin_slot": 2, + "target_id": 92, + "target_slot": 0, + "type": "*" + }, + { + "id": 114, + "origin_id": -10, + "origin_slot": 3, + "target_id": 93, + "target_slot": 0, + "type": "*" + }, + { + "id": 115, + "origin_id": 83, + "origin_slot": 0, + "target_id": 94, + "target_slot": 0, + "type": "STRING" + }, + { + "id": 116, + "origin_id": 92, + "origin_slot": 0, + "target_id": 94, + "target_slot": 2, + "type": "STRING" + }, + { + "id": 117, + "origin_id": 94, + "origin_slot": 0, + "target_id": 95, + "target_slot": 0, + "type": "STRING" + }, + { + "id": 118, + "origin_id": 93, + "origin_slot": 0, + "target_id": 95, + "target_slot": 2, + "type": "STRING" + }, + { + "id": 119, + "origin_id": 95, + "origin_slot": 0, + "target_id": 74, + "target_slot": 2, + "type": "STRING" + }, + { + "id": 120, + "origin_id": -10, + "origin_slot": 9, + "target_id": 91, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 121, + "origin_id": -10, + "origin_slot": 10, + "target_id": 63, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 124, + "origin_id": 75, + "origin_slot": 0, + "target_id": 97, + "target_slot": 0, + "type": "STRING" + } + ], + "extra": {}, + "category": "Image generation and editing/Text to image", + "description": "Generates images from text prompts using Baidu’s open ERNIE Image (~8B DiT): bilingual in-image typography and layouts (posters, infographics, multi-panel compositions) alongside general scenes, with bundled encoders and VAE." + } + ] + }, + "extra": { + "ue_links": [] + } +} diff --git a/blueprints/Text to Image (Flux.1 Dev).json b/blueprints/Text to Image (Flux.1 Dev).json index 45f68f508..6d8446e81 100644 --- a/blueprints/Text to Image (Flux.1 Dev).json +++ b/blueprints/Text to Image (Flux.1 Dev).json @@ -1030,7 +1030,7 @@ "workflowRendererVersion": "LG" }, "category": "Image generation and editing/Text to image", - "description": "Generates images from text prompts using Flux.1 [dev], Black Forest Labs' 12B diffusion model." + "description": "Generates images from prompts using FLUX.1 [dev]: a 12B rectified-flow MMDiT with dual CLIP plus T5-XXL text encoders and guidance-distilled sampling for sharp prompt following versus classic DDPM diffusion." } ] }, diff --git a/blueprints/Text to Image (Flux.1 Krea Dev).json b/blueprints/Text to Image (Flux.1 Krea Dev).json index 30a78dca1..0d7fa03c4 100644 --- a/blueprints/Text to Image (Flux.1 Krea Dev).json +++ b/blueprints/Text to Image (Flux.1 Krea Dev).json @@ -1024,7 +1024,7 @@ "workflowRendererVersion": "LG" }, "category": "Image generation and editing/Text to image", - "description": "Generates images from text prompts using Flux.1 Krea Dev, a Black Forest Labs × Krea collaboration variant." + "description": "FLUX.1 Krea [dev] (Black Forest Labs × Krea): open-weight 12B rectified-flow text-to-image drop-in alongside FLUX.1 [dev], tuned away from overcooked saturation toward more natural diversity in people, realism, and style while keeping ecosystem compatibility." } ] }, diff --git a/blueprints/Text to Image (Flux.2 Dev).json b/blueprints/Text to Image (Flux.2 Dev).json new file mode 100644 index 000000000..d5ca3077d --- /dev/null +++ b/blueprints/Text to Image (Flux.2 Dev).json @@ -0,0 +1,1870 @@ +{ + "revision": 0, + "last_node_id": 123, + "last_link_id": 0, + "nodes": [ + { + "id": 123, + "type": "85066daf-feda-4c7b-bbc3-d4797e8ccf0f", + "pos": [ + -800, + 640 + ], + "size": [ + 400, + 0 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "label": "prompt", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": null + }, + { + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": null + }, + { + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": null + }, + { + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": null + }, + { + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": null + }, + { + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": null + }, + { + "label": "turbo_lora", + "name": "lora_name", + "type": "COMBO", + "widget": { + "name": "lora_name" + }, + "link": null + }, + { + "label": "enable_turbo_mode", + "name": "value", + "type": "BOOLEAN", + "widget": { + "name": "value" + }, + "link": null + }, + { + "name": "noise_seed", + "type": "INT", + "widget": { + "name": "noise_seed" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [] + } + ], + "properties": { + "proxyWidgets": [ + [ + "115", + "text" + ], + [ + "113", + "width" + ], + [ + "113", + "height" + ], + [ + "122", + "unet_name" + ], + [ + "111", + "clip_name" + ], + [ + "108", + "vae_name" + ], + [ + "116", + "lora_name" + ], + [ + "121", + "value" + ], + [ + "114", + "noise_seed" + ], + [ + "114", + "control_after_generate" + ] + ], + "cnr_id": "comfy-core", + "ver": "0.15.1", + "ue_properties": { + "widget_ue_connectable": { + "value": true, + "lora_name": true + }, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [], + "title": "Text to Image (Flux.2 Dev)" + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "85066daf-feda-4c7b-bbc3-d4797e8ccf0f", + "version": 1, + "state": { + "lastGroupId": 6, + "lastNodeId": 123, + "lastLinkId": 232, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "Text to Image (Flux.2 Dev)", + "inputNode": { + "id": -10, + "bounding": [ + -1500, + 250, + 151.744140625, + 220 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + 1560, + -20, + 120, + 60 + ] + }, + "inputs": [ + { + "id": "1f4f1091-3f97-41d8-8ed8-e8b02260cf3c", + "name": "text", + "type": "STRING", + "linkIds": [ + 206 + ], + "label": "prompt", + "pos": [ + -1368.255859375, + 270 + ] + }, + { + "id": "b9b59411-4f5f-4482-8f78-369e6d50e71c", + "name": "width", + "type": "INT", + "linkIds": [ + 222, + 231 + ], + "pos": [ + -1368.255859375, + 290 + ] + }, + { + "id": "c6de9a28-3bf6-40d0-be16-f75ec517a766", + "name": "height", + "type": "INT", + "linkIds": [ + 223, + 232 + ], + "pos": [ + -1368.255859375, + 310 + ] + }, + { + "id": "8f1b1c75-e47c-45f5-af57-74abcfe8967c", + "name": "unet_name", + "type": "COMBO", + "linkIds": [ + 225 + ], + "pos": [ + -1368.255859375, + 330 + ] + }, + { + "id": "6ac27631-1bf0-4161-9670-a662f6180b94", + "name": "clip_name", + "type": "COMBO", + "linkIds": [ + 226 + ], + "pos": [ + -1368.255859375, + 350 + ] + }, + { + "id": "932e6cbe-f716-4905-ae54-d2b3543497bd", + "name": "vae_name", + "type": "COMBO", + "linkIds": [ + 227 + ], + "pos": [ + -1368.255859375, + 370 + ] + }, + { + "id": "37400048-5e7b-427b-8b79-ea35841d5306", + "name": "lora_name", + "type": "COMBO", + "linkIds": [ + 228 + ], + "label": "turbo_lora", + "pos": [ + -1368.255859375, + 390 + ] + }, + { + "id": "333212d0-f027-476f-8b97-a921e20e340a", + "name": "value", + "type": "BOOLEAN", + "linkIds": [ + 229 + ], + "label": "enable_turbo_mode", + "pos": [ + -1368.255859375, + 410 + ] + }, + { + "id": "e7e73fad-ce6e-48d5-b719-e2abed685185", + "name": "noise_seed", + "type": "INT", + "linkIds": [ + 230 + ], + "pos": [ + -1368.255859375, + 430 + ] + } + ], + "outputs": [ + { + "id": "ed3c0a0f-a39f-453e-907f-8249c8e3335d", + "name": "IMAGE", + "type": "IMAGE", + "linkIds": [ + 9 + ], + "localized_name": "IMAGE", + "pos": [ + 1580, + 0 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 105, + "type": "BasicGuider", + "pos": [ + 570, + 170 + ], + "size": [ + 230, + 100 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 210 + }, + { + "localized_name": "conditioning", + "name": "conditioning", + "type": "CONDITIONING", + "link": 165 + } + ], + "outputs": [ + { + "localized_name": "GUIDER", + "name": "GUIDER", + "type": "GUIDER", + "slot_index": 0, + "links": [ + 30 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "BasicGuider", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 106, + "type": "FluxGuidance", + "pos": [ + -510, + 470 + ], + "size": [ + 320, + 110 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [ + { + "localized_name": "conditioning", + "name": "conditioning", + "type": "CONDITIONING", + "link": 41 + }, + { + "localized_name": "guidance", + "name": "guidance", + "type": "FLOAT", + "widget": { + "name": "guidance" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "slot_index": 0, + "links": [ + 165 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "FluxGuidance", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 4 + ], + "color": "#233", + "bgcolor": "#355" + }, + { + "id": 107, + "type": "KSamplerSelect", + "pos": [ + 570, + 350 + ], + "size": [ + 230, + 110 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "localized_name": "sampler_name", + "name": "sampler_name", + "type": "COMBO", + "widget": { + "name": "sampler_name" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "SAMPLER", + "name": "SAMPLER", + "type": "SAMPLER", + "links": [ + 19 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "KSamplerSelect", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "euler" + ] + }, + { + "id": 108, + "type": "VAELoader", + "pos": [ + -1000, + 460 + ], + "size": [ + 300, + 110 + ], + "flags": {}, + "order": 5, + "mode": 0, + "inputs": [ + { + "localized_name": "vae_name", + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": 227 + } + ], + "outputs": [ + { + "localized_name": "VAE", + "name": "VAE", + "type": "VAE", + "slot_index": 0, + "links": [ + 159 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "VAELoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "full_encoder_small_decoder.safetensors", + "url": "https://huggingface.co/black-forest-labs/FLUX.2-small-decoder/resolve/main/full_encoder_small_decoder.safetensors", + "directory": "vae" + } + ] + }, + "widgets_values": [ + "full_encoder_small_decoder.safetensors" + ] + }, + { + "id": 109, + "type": "SamplerCustomAdvanced", + "pos": [ + 860, + -20 + ], + "size": [ + 280, + 330 + ], + "flags": {}, + "order": 6, + "mode": 0, + "inputs": [ + { + "localized_name": "noise", + "name": "noise", + "type": "NOISE", + "link": 37 + }, + { + "localized_name": "guider", + "name": "guider", + "type": "GUIDER", + "link": 30 + }, + { + "localized_name": "sampler", + "name": "sampler", + "type": "SAMPLER", + "link": 19 + }, + { + "localized_name": "sigmas", + "name": "sigmas", + "type": "SIGMAS", + "link": 132 + }, + { + "localized_name": "latent_image", + "name": "latent_image", + "type": "LATENT", + "link": 161 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "LATENT", + "slot_index": 0, + "links": [ + 24 + ] + }, + { + "localized_name": "denoised_output", + "name": "denoised_output", + "type": "LATENT", + "links": null + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "SamplerCustomAdvanced", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 110, + "type": "VAEDecode", + "pos": [ + 1220, + -20 + ], + "size": [ + 230, + 100 + ], + "flags": {}, + "order": 7, + "mode": 0, + "inputs": [ + { + "localized_name": "samples", + "name": "samples", + "type": "LATENT", + "link": 24 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 159 + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "slot_index": 0, + "links": [ + 9 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "VAEDecode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 111, + "type": "CLIPLoader", + "pos": [ + -1000, + 200 + ], + "size": [ + 300, + 150 + ], + "flags": {}, + "order": 8, + "mode": 0, + "inputs": [ + { + "localized_name": "clip_name", + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": 226 + }, + { + "localized_name": "type", + "name": "type", + "type": "COMBO", + "widget": { + "name": "type" + }, + "link": null + }, + { + "localized_name": "device", + "name": "device", + "shape": 7, + "type": "COMBO", + "widget": { + "name": "device" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [ + 117 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "CLIPLoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "mistral_3_small_flux2_bf16.safetensors", + "url": "https://huggingface.co/Comfy-Org/flux2-dev/resolve/main/split_files/text_encoders/mistral_3_small_flux2_bf16.safetensors", + "directory": "text_encoders" + } + ] + }, + "widgets_values": [ + "mistral_3_small_flux2_bf16.safetensors", + "flux2", + "default" + ] + }, + { + "id": 112, + "type": "Flux2Scheduler", + "pos": [ + 570, + 550 + ], + "size": [ + 230, + 170 + ], + "flags": {}, + "order": 9, + "mode": 0, + "inputs": [ + { + "localized_name": "steps", + "name": "steps", + "type": "INT", + "widget": { + "name": "steps" + }, + "link": 213 + }, + { + "localized_name": "width", + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": 231 + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": 232 + } + ], + "outputs": [ + { + "localized_name": "SIGMAS", + "name": "SIGMAS", + "type": "SIGMAS", + "links": [ + 132 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "Flux2Scheduler", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 20, + 1024, + 1024 + ] + }, + { + "id": 113, + "type": "EmptyFlux2LatentImage", + "pos": [ + -980, + 660 + ], + "size": [ + 270, + 170 + ], + "flags": {}, + "order": 10, + "mode": 0, + "inputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": 222 + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": 223 + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "widget": { + "name": "batch_size" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "links": [ + 161 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "EmptyFlux2LatentImage", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 1024, + 1024, + 1 + ] + }, + { + "id": 114, + "type": "RandomNoise", + "pos": [ + 570, + -20 + ], + "size": [ + 230, + 110 + ], + "flags": {}, + "order": 11, + "mode": 0, + "inputs": [ + { + "localized_name": "noise_seed", + "name": "noise_seed", + "type": "INT", + "widget": { + "name": "noise_seed" + }, + "link": 230 + } + ], + "outputs": [ + { + "localized_name": "NOISE", + "name": "NOISE", + "type": "NOISE", + "links": [ + 37 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "RandomNoise", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 1027111520328378, + "randomize" + ] + }, + { + "id": 115, + "type": "CLIPTextEncode", + "pos": [ + -630, + -40 + ], + "size": [ + 440, + 450 + ], + "flags": {}, + "order": 12, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 117 + }, + { + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": 206 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "slot_index": 0, + "links": [ + 41 + ] + } + ], + "title": "CLIP Text Encode (Positive Prompt)", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "CLIPTextEncode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "" + ], + "color": "#232", + "bgcolor": "#353" + }, + { + "id": 116, + "type": "LoraLoaderModelOnly", + "pos": [ + -150, + 220 + ], + "size": [ + 300, + 140 + ], + "flags": {}, + "order": 13, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 221 + }, + { + "localized_name": "lora_name", + "name": "lora_name", + "type": "COMBO", + "widget": { + "name": "lora_name" + }, + "link": 228 + }, + { + "localized_name": "strength_model", + "name": "strength_model", + "type": "FLOAT", + "widget": { + "name": "strength_model" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 209 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.7.0", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "LoraLoaderModelOnly", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "Flux_2-Turbo-LoRA_comfyui.safetensors", + "url": "https://huggingface.co/ByteZSzn/Flux.2-Turbo-ComfyUI/resolve/main/Flux_2-Turbo-LoRA_comfyui.safetensors", + "directory": "loras" + } + ] + }, + "widgets_values": [ + "Flux_2-Turbo-LoRA_comfyui.safetensors", + 1 + ] + }, + { + "id": 117, + "type": "ComfySwitchNode", + "pos": [ + 220, + -30 + ], + "size": [ + 270, + 130 + ], + "flags": {}, + "order": 14, + "mode": 0, + "inputs": [ + { + "localized_name": "on_false", + "name": "on_false", + "type": "*", + "link": 208 + }, + { + "localized_name": "on_true", + "name": "on_true", + "type": "*", + "link": 209 + }, + { + "localized_name": "switch", + "name": "switch", + "type": "BOOLEAN", + "widget": { + "name": "switch" + }, + "link": 215 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "*", + "links": [ + 210 + ] + } + ], + "title": "Switch(model)", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.15.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ComfySwitchNode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + false + ] + }, + { + "id": 118, + "type": "PrimitiveInt", + "pos": [ + -140, + -30 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 211 + ] + } + ], + "title": "Steps", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.15.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "PrimitiveInt", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 20, + "fixed" + ] + }, + { + "id": 119, + "type": "PrimitiveInt", + "pos": [ + -150, + 460 + ], + "size": [ + 300, + 110 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "INT", + "widget": { + "name": "value" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 212 + ] + } + ], + "title": "Steps", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.15.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "PrimitiveInt", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 8, + "fixed" + ] + }, + { + "id": 120, + "type": "ComfySwitchNode", + "pos": [ + 220, + 260 + ], + "size": [ + 270, + 130 + ], + "flags": {}, + "order": 15, + "mode": 0, + "inputs": [ + { + "localized_name": "on_false", + "name": "on_false", + "type": "*", + "link": 211 + }, + { + "localized_name": "on_true", + "name": "on_true", + "type": "*", + "link": 212 + }, + { + "localized_name": "switch", + "name": "switch", + "type": "BOOLEAN", + "widget": { + "name": "switch" + }, + "link": 214 + } + ], + "outputs": [ + { + "localized_name": "output", + "name": "output", + "type": "*", + "links": [ + 213 + ] + } + ], + "title": "Switch(steps)", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.15.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ComfySwitchNode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + false + ] + }, + { + "id": 121, + "type": "PrimitiveBoolean", + "pos": [ + -110, + 690 + ], + "size": [ + 270, + 100 + ], + "flags": {}, + "order": 16, + "mode": 0, + "inputs": [ + { + "localized_name": "value", + "name": "value", + "type": "BOOLEAN", + "widget": { + "name": "value" + }, + "link": 229 + } + ], + "outputs": [ + { + "localized_name": "BOOLEAN", + "name": "BOOLEAN", + "type": "BOOLEAN", + "links": [ + 214, + 215 + ] + } + ], + "title": "Enable Turbo LoRA", + "properties": { + "cnr_id": "comfy-core", + "ver": "0.15.1", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "PrimitiveBoolean", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + false + ] + }, + { + "id": 122, + "type": "UNETLoader", + "pos": [ + -1000, + -30 + ], + "size": [ + 300, + 110 + ], + "flags": {}, + "order": 17, + "mode": 0, + "inputs": [ + { + "localized_name": "unet_name", + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": 225 + }, + { + "localized_name": "weight_dtype", + "name": "weight_dtype", + "type": "COMBO", + "widget": { + "name": "weight_dtype" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "slot_index": 0, + "links": [ + 208, + 221 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.71", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "UNETLoader", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "flux2_dev_fp8mixed.safetensors", + "url": "https://huggingface.co/Comfy-Org/flux2-dev/resolve/main/split_files/diffusion_models/flux2_dev_fp8mixed.safetensors", + "directory": "diffusion_models" + } + ] + }, + "widgets_values": [ + "flux2_dev_fp8mixed.safetensors", + "default" + ] + } + ], + "groups": [ + { + "id": 1, + "title": "Step 1 - Upload models", + "bounding": [ + -1040, + -110, + 380, + 710 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 2, + "title": "Custom sampler", + "bounding": [ + 540, + -110, + 640, + 870 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 4, + "title": "Step2 - Prompt", + "bounding": [ + -640, + -110, + 460, + 710 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 5, + "title": "Original", + "bounding": [ + -160, + -110, + 320, + 230 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 6, + "title": "8 Steps LoRA", + "bounding": [ + -160, + 140, + 320, + 460 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + } + ], + "links": [ + { + "id": 165, + "origin_id": 106, + "origin_slot": 0, + "target_id": 105, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 41, + "origin_id": 115, + "origin_slot": 0, + "target_id": 106, + "target_slot": 0, + "type": "CONDITIONING" + }, + { + "id": 37, + "origin_id": 114, + "origin_slot": 0, + "target_id": 109, + "target_slot": 0, + "type": "NOISE" + }, + { + "id": 30, + "origin_id": 105, + "origin_slot": 0, + "target_id": 109, + "target_slot": 1, + "type": "GUIDER" + }, + { + "id": 19, + "origin_id": 107, + "origin_slot": 0, + "target_id": 109, + "target_slot": 2, + "type": "SAMPLER" + }, + { + "id": 132, + "origin_id": 112, + "origin_slot": 0, + "target_id": 109, + "target_slot": 3, + "type": "SIGMAS" + }, + { + "id": 161, + "origin_id": 113, + "origin_slot": 0, + "target_id": 109, + "target_slot": 4, + "type": "LATENT" + }, + { + "id": 117, + "origin_id": 111, + "origin_slot": 0, + "target_id": 115, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 24, + "origin_id": 109, + "origin_slot": 0, + "target_id": 110, + "target_slot": 0, + "type": "LATENT" + }, + { + "id": 159, + "origin_id": 108, + "origin_slot": 0, + "target_id": 110, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 9, + "origin_id": 110, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 206, + "origin_id": -10, + "origin_slot": 0, + "target_id": 115, + "target_slot": 1, + "type": "STRING" + }, + { + "id": 208, + "origin_id": 122, + "origin_slot": 0, + "target_id": 117, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 209, + "origin_id": 116, + "origin_slot": 0, + "target_id": 117, + "target_slot": 1, + "type": "MODEL" + }, + { + "id": 210, + "origin_id": 117, + "origin_slot": 0, + "target_id": 105, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 211, + "origin_id": 118, + "origin_slot": 0, + "target_id": 120, + "target_slot": 0, + "type": "INT" + }, + { + "id": 212, + "origin_id": 119, + "origin_slot": 0, + "target_id": 120, + "target_slot": 1, + "type": "INT" + }, + { + "id": 213, + "origin_id": 120, + "origin_slot": 0, + "target_id": 112, + "target_slot": 0, + "type": "INT" + }, + { + "id": 214, + "origin_id": 121, + "origin_slot": 0, + "target_id": 120, + "target_slot": 2, + "type": "BOOLEAN" + }, + { + "id": 215, + "origin_id": 121, + "origin_slot": 0, + "target_id": 117, + "target_slot": 2, + "type": "BOOLEAN" + }, + { + "id": 221, + "origin_id": 122, + "origin_slot": 0, + "target_id": 116, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 222, + "origin_id": -10, + "origin_slot": 1, + "target_id": 113, + "target_slot": 0, + "type": "INT" + }, + { + "id": 223, + "origin_id": -10, + "origin_slot": 2, + "target_id": 113, + "target_slot": 1, + "type": "INT" + }, + { + "id": 225, + "origin_id": -10, + "origin_slot": 3, + "target_id": 122, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 226, + "origin_id": -10, + "origin_slot": 4, + "target_id": 111, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 227, + "origin_id": -10, + "origin_slot": 5, + "target_id": 108, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 228, + "origin_id": -10, + "origin_slot": 6, + "target_id": 116, + "target_slot": 1, + "type": "COMBO" + }, + { + "id": 229, + "origin_id": -10, + "origin_slot": 7, + "target_id": 121, + "target_slot": 0, + "type": "BOOLEAN" + }, + { + "id": 230, + "origin_id": -10, + "origin_slot": 8, + "target_id": 114, + "target_slot": 0, + "type": "INT" + }, + { + "id": 231, + "origin_id": -10, + "origin_slot": 1, + "target_id": 112, + "target_slot": 1, + "type": "INT" + }, + { + "id": 232, + "origin_id": -10, + "origin_slot": 2, + "target_id": 112, + "target_slot": 2, + "type": "INT" + } + ], + "extra": { + "workflowRendererVersion": "LG" + }, + "category": "Image generation and editing/Text to image", + "description": "Generates images from prompts using FLUX.2 [dev]: a newer 32B rectified-flow stack with distilled guidance plus a stronger long-context multimodal encoder for complex scenes, sharper typography/UI text, anatomy, lighting, and high-resolution latent decoding." + } + ] + }, + "extra": { + "ue_links": [] + } +} diff --git a/blueprints/Text to Image (Z-Image-Base).json b/blueprints/Text to Image (Z-Image-Base).json new file mode 100644 index 000000000..169263712 --- /dev/null +++ b/blueprints/Text to Image (Z-Image-Base).json @@ -0,0 +1,1184 @@ +{ + "revision": 0, + "last_node_id": 126, + "last_link_id": 0, + "nodes": [ + { + "id": 126, + "type": "8a2bb267-5858-4aaf-bdcd-61002711af19", + "pos": [ + -2280, + 2850 + ], + "size": [ + 410, + 560 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "label": "prompt", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": null + }, + { + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": null + }, + { + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": null + }, + { + "name": "steps", + "type": "INT", + "widget": { + "name": "steps" + }, + "link": null + }, + { + "name": "cfg", + "type": "FLOAT", + "widget": { + "name": "cfg" + }, + "link": null + }, + { + "name": "seed", + "type": "INT", + "widget": { + "name": "seed" + }, + "link": null + }, + { + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": null + }, + { + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": null + }, + { + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [] + } + ], + "properties": { + "proxyWidgets": [ + [ + "67", + "text" + ], + [ + "68", + "width" + ], + [ + "68", + "height" + ], + [ + "69", + "steps" + ], + [ + "69", + "cfg" + ], + [ + "69", + "seed" + ], + [ + "66", + "unet_name" + ], + [ + "62", + "clip_name" + ], + [ + "63", + "vae_name" + ], + [ + "69", + "control_after_generate" + ] + ], + "cnr_id": "comfy-core", + "ver": "0.13.0", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [], + "title": "Text to Image (Z-Image-Base)" + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "8a2bb267-5858-4aaf-bdcd-61002711af19", + "version": 1, + "state": { + "lastGroupId": 16, + "lastNodeId": 126, + "lastLinkId": 229, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "Text to Image (Z-Image-Base)", + "description": "Generates images from text prompts using Z-Image base weights with Qwen3 text encoder and bundled VAE.", + "inputNode": { + "id": -10, + "bounding": [ + -220, + 40, + 120, + 220 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + 1840, + -150, + 120, + 60 + ] + }, + "inputs": [ + { + "id": "af36fee5-4f8b-4a8e-bfa8-cb8fe7006cc3", + "name": "text", + "type": "STRING", + "linkIds": [ + 108 + ], + "label": "prompt", + "pos": [ + -120, + 60 + ] + }, + { + "id": "357f0059-e8e6-41f6-a290-c53b0a60c0ed", + "name": "width", + "type": "INT", + "linkIds": [ + 114 + ], + "pos": [ + -120, + 80 + ] + }, + { + "id": "4a442743-a9c2-4aa5-9efd-05d43f3322d3", + "name": "height", + "type": "INT", + "linkIds": [ + 115 + ], + "pos": [ + -120, + 100 + ] + }, + { + "id": "a0fc336b-d349-418e-8415-318653f7b6b3", + "name": "steps", + "type": "INT", + "linkIds": [ + 116 + ], + "pos": [ + -120, + 120 + ] + }, + { + "id": "2f253ace-1e1a-415f-9b95-a10430bd5749", + "name": "cfg", + "type": "FLOAT", + "linkIds": [ + 117 + ], + "pos": [ + -120, + 140 + ] + }, + { + "id": "18a6ad37-23aa-4bf7-a0cd-1d6ca6e2a128", + "name": "seed", + "type": "INT", + "linkIds": [ + 118 + ], + "pos": [ + -120, + 160 + ] + }, + { + "id": "d1fc4937-8505-4ec6-9fc4-a33ef7b45eee", + "name": "unet_name", + "type": "COMBO", + "linkIds": [ + 119 + ], + "pos": [ + -120, + 180 + ] + }, + { + "id": "db45dd49-d990-4ceb-a849-f96341874cdd", + "name": "clip_name", + "type": "COMBO", + "linkIds": [ + 120 + ], + "pos": [ + -120, + 200 + ] + }, + { + "id": "37b8eac6-9b1b-452b-81f3-0ba9e34a576a", + "name": "vae_name", + "type": "COMBO", + "linkIds": [ + 121 + ], + "pos": [ + -120, + 220 + ] + } + ], + "outputs": [ + { + "id": "f2bea309-bfe7-4ccb-9ffe-9475bf1da2ae", + "name": "IMAGE", + "type": "IMAGE", + "linkIds": [ + 79 + ], + "localized_name": "IMAGE", + "pos": [ + 1860, + -130 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 67, + "type": "CLIPTextEncode", + "pos": [ + 600, + -90 + ], + "size": [ + 410, + 320 + ], + "flags": {}, + "order": 5, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 78 + }, + { + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": 108 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 75 + ] + } + ], + "title": "CLIP Text Encode (Positive Prompt)", + "properties": { + "Node name for S&R": "CLIPTextEncode", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "" + ], + "color": "#232", + "bgcolor": "#353" + }, + { + "id": 68, + "type": "EmptySD3LatentImage", + "pos": [ + 240, + 620 + ], + "size": [ + 260, + 170 + ], + "flags": {}, + "order": 6, + "mode": 0, + "inputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": 114 + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": 115 + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "widget": { + "name": "batch_size" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "slot_index": 0, + "links": [ + 77 + ] + } + ], + "properties": { + "Node name for S&R": "EmptySD3LatentImage", + "cnr_id": "comfy-core", + "ver": "0.3.64", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 1024, + 1024, + 1 + ] + }, + { + "id": 63, + "type": "VAELoader", + "pos": [ + 230, + 340 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "vae_name", + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": 121 + } + ], + "outputs": [ + { + "localized_name": "VAE", + "name": "VAE", + "type": "VAE", + "links": [ + 73 + ] + } + ], + "properties": { + "Node name for S&R": "VAELoader", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "models": [ + { + "name": "ae.safetensors", + "url": "https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/vae/ae.safetensors", + "directory": "vae" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "ae.safetensors" + ] + }, + { + "id": 62, + "type": "CLIPLoader", + "pos": [ + 230, + 110 + ], + "size": [ + 270, + 150 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "clip_name", + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": 120 + }, + { + "localized_name": "type", + "name": "type", + "type": "COMBO", + "widget": { + "name": "type" + }, + "link": null + }, + { + "localized_name": "device", + "name": "device", + "shape": 7, + "type": "COMBO", + "widget": { + "name": "device" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [ + 78, + 82 + ] + } + ], + "properties": { + "Node name for S&R": "CLIPLoader", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "models": [ + { + "name": "qwen_3_4b.safetensors", + "url": "https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/text_encoders/qwen_3_4b.safetensors", + "directory": "text_encoders" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "qwen_3_4b.safetensors", + "lumina2", + "default" + ] + }, + { + "id": 65, + "type": "VAEDecode", + "pos": [ + 1450, + -150 + ], + "size": [ + 230, + 100 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "samples", + "name": "samples", + "type": "LATENT", + "link": 72 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 73 + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "slot_index": 0, + "links": [ + 79 + ] + } + ], + "properties": { + "Node name for S&R": "VAEDecode", + "cnr_id": "comfy-core", + "ver": "0.3.64", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 70, + "type": "ModelSamplingAuraFlow", + "pos": [ + 1100, + -150 + ], + "size": [ + 310, + 110 + ], + "flags": {}, + "order": 8, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 109 + }, + { + "localized_name": "shift", + "name": "shift", + "type": "FLOAT", + "widget": { + "name": "shift" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "slot_index": 0, + "links": [ + 74 + ] + } + ], + "properties": { + "Node name for S&R": "ModelSamplingAuraFlow", + "cnr_id": "comfy-core", + "ver": "0.3.64", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 3 + ] + }, + { + "id": 66, + "type": "UNETLoader", + "pos": [ + 230, + -90 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [ + { + "localized_name": "unet_name", + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": 119 + }, + { + "localized_name": "weight_dtype", + "name": "weight_dtype", + "type": "COMBO", + "widget": { + "name": "weight_dtype" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 109 + ] + } + ], + "properties": { + "Node name for S&R": "UNETLoader", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "models": [ + { + "name": "z_image_bf16.safetensors", + "url": "https://huggingface.co/Comfy-Org/z_image/resolve/main/split_files/diffusion_models/z_image_bf16.safetensors", + "directory": "diffusion_models" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "z_image_bf16.safetensors", + "default" + ] + }, + { + "id": 71, + "type": "CLIPTextEncode", + "pos": [ + 600, + 310 + ], + "size": [ + 390, + 140 + ], + "flags": {}, + "order": 9, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 82 + }, + { + "label": "prompt", + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 83 + ] + } + ], + "title": "CLIP Text Encode (Negative Prompt)", + "properties": { + "Node name for S&R": "CLIPTextEncode", + "cnr_id": "comfy-core", + "ver": "0.3.73", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "" + ], + "color": "#323", + "bgcolor": "#535" + }, + { + "id": 69, + "type": "KSampler", + "pos": [ + 1100, + 10 + ], + "size": [ + 310, + 440 + ], + "flags": {}, + "order": 7, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 74 + }, + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "link": 75 + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "link": 83 + }, + { + "localized_name": "latent_image", + "name": "latent_image", + "type": "LATENT", + "link": 77 + }, + { + "localized_name": "seed", + "name": "seed", + "type": "INT", + "widget": { + "name": "seed" + }, + "link": 118 + }, + { + "localized_name": "steps", + "name": "steps", + "type": "INT", + "widget": { + "name": "steps" + }, + "link": 116 + }, + { + "localized_name": "cfg", + "name": "cfg", + "type": "FLOAT", + "widget": { + "name": "cfg" + }, + "link": 117 + }, + { + "localized_name": "sampler_name", + "name": "sampler_name", + "type": "COMBO", + "widget": { + "name": "sampler_name" + }, + "link": null + }, + { + "localized_name": "scheduler", + "name": "scheduler", + "type": "COMBO", + "widget": { + "name": "scheduler" + }, + "link": null + }, + { + "localized_name": "denoise", + "name": "denoise", + "type": "FLOAT", + "widget": { + "name": "denoise" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "slot_index": 0, + "links": [ + 72 + ] + } + ], + "properties": { + "Node name for S&R": "KSampler", + "cnr_id": "comfy-core", + "ver": "0.3.64", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 0, + "randomize", + 25, + 4, + "res_multistep", + "simple", + 1 + ] + }, + { + "id": 87, + "type": "MarkdownNote", + "pos": [ + 1110, + -360 + ], + "size": [ + 300, + 120 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [], + "outputs": [], + "properties": {}, + "widgets_values": [ + "- Steps: 30～50\n- cfg: 3～5" + ], + "color": "#222", + "bgcolor": "#000", + "title": "Original Settings" + } + ], + "groups": [ + { + "id": 2, + "title": "Step2 - Image size", + "bounding": [ + 200, + 530, + 330, + 287.9999544955691 + ], + "color": "#3f789e", + "flags": {} + }, + { + "id": 3, + "title": "Step3 - Prompt", + "bounding": [ + 570, + -200, + 470, + 700 + ], + "color": "#3f789e", + "flags": {} + }, + { + "id": 4, + "title": "Step1 - Load models", + "bounding": [ + 200, + -200, + 330, + 700 + ], + "color": "#3f789e", + "flags": {} + } + ], + "links": [ + { + "id": 78, + "origin_id": 62, + "origin_slot": 0, + "target_id": 67, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 74, + "origin_id": 70, + "origin_slot": 0, + "target_id": 69, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 75, + "origin_id": 67, + "origin_slot": 0, + "target_id": 69, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 83, + "origin_id": 71, + "origin_slot": 0, + "target_id": 69, + "target_slot": 2, + "type": "CONDITIONING" + }, + { + "id": 77, + "origin_id": 68, + "origin_slot": 0, + "target_id": 69, + "target_slot": 3, + "type": "LATENT" + }, + { + "id": 82, + "origin_id": 62, + "origin_slot": 0, + "target_id": 71, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 72, + "origin_id": 69, + "origin_slot": 0, + "target_id": 65, + "target_slot": 0, + "type": "LATENT" + }, + { + "id": 73, + "origin_id": 63, + "origin_slot": 0, + "target_id": 65, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 79, + "origin_id": 65, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 108, + "origin_id": -10, + "origin_slot": 0, + "target_id": 67, + "target_slot": 1, + "type": "STRING" + }, + { + "id": 109, + "origin_id": 66, + "origin_slot": 0, + "target_id": 70, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 114, + "origin_id": -10, + "origin_slot": 1, + "target_id": 68, + "target_slot": 0, + "type": "INT" + }, + { + "id": 115, + "origin_id": -10, + "origin_slot": 2, + "target_id": 68, + "target_slot": 1, + "type": "INT" + }, + { + "id": 116, + "origin_id": -10, + "origin_slot": 3, + "target_id": 69, + "target_slot": 5, + "type": "INT" + }, + { + "id": 117, + "origin_id": -10, + "origin_slot": 4, + "target_id": 69, + "target_slot": 6, + "type": "FLOAT" + }, + { + "id": 118, + "origin_id": -10, + "origin_slot": 5, + "target_id": 69, + "target_slot": 4, + "type": "INT" + }, + { + "id": 119, + "origin_id": -10, + "origin_slot": 6, + "target_id": 66, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 120, + "origin_id": -10, + "origin_slot": 7, + "target_id": 62, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 121, + "origin_id": -10, + "origin_slot": 8, + "target_id": 63, + "target_slot": 0, + "type": "COMBO" + } + ], + "extra": { + "workflowRendererVersion": "LG" + }, + "category": "Image generation and editing/Text to image" + } + ] + }, + "extra": {} +} \ No newline at end of file diff --git a/blueprints/Text to Image (Z-Image-Turbo).json b/blueprints/Text to Image (Z-Image-Turbo).json index 6975151ea..2501486fa 100644 --- a/blueprints/Text to Image (Z-Image-Turbo).json +++ b/blueprints/Text to Image (Z-Image-Turbo).json @@ -1,22 +1,21 @@ { - "id": "1c3eaa76-5cfa-4dc7-8571-97a570324e01", "revision": 0, - "last_node_id": 34, - "last_link_id": 40, + "last_node_id": 57, + "last_link_id": 0, "nodes": [ { - "id": 5, - "type": "dfe9eb32-97c0-43a5-90d5-4fd37768d91b", + "id": 57, + "type": "f2fdebf6-dfaf-43b6-9eb2-7f70613cfdc1", "pos": [ - -2.5766491043910378e-05, - 1229.999928629805 + 130, + 200 ], "size": [ 400, 470 ], "flags": {}, - "order": 0, + "order": 1, "mode": 0, "inputs": [ { @@ -44,6 +43,22 @@ }, "link": null }, + { + "name": "seed", + "type": "INT", + "widget": { + "name": "seed" + }, + "link": null + }, + { + "name": "steps", + "type": "INT", + "widget": { + "name": "steps" + }, + "link": null + }, { "name": "unet_name", "type": "COMBO", @@ -80,15 +95,15 @@ "properties": { "proxyWidgets": [ [ - "-1", + "27", "text" ], [ - "-1", + "13", "width" ], [ - "-1", + "13", "height" ], [ @@ -97,19 +112,23 @@ ], [ "3", - "control_after_generate" + "steps" ], [ - "-1", + "28", "unet_name" ], [ - "-1", + "30", "clip_name" ], [ - "-1", + "29", "vae_name" + ], + [ + "3", + "control_after_generate" ] ], "cnr_id": "comfy-core", @@ -122,29 +141,21 @@ "secondTabOffset": 80, "secondTabWidth": 65 }, - "widgets_values": [ - "", - 1024, - 1024, - null, - null, - "z_image_turbo_bf16.safetensors", - "qwen_3_4b.safetensors", - "ae.safetensors" - ] + "widgets_values": [], + "title": "Text to Image (Z-Image-Turbo)" } ], "links": [], - "groups": [], + "version": 0.4, "definitions": { "subgraphs": [ { - "id": "dfe9eb32-97c0-43a5-90d5-4fd37768d91b", + "id": "f2fdebf6-dfaf-43b6-9eb2-7f70613cfdc1", "version": 1, "state": { "lastGroupId": 4, - "lastNodeId": 34, - "lastLinkId": 40, + "lastNodeId": 61, + "lastLinkId": 75, "lastRerouteId": 0 }, "revision": 0, @@ -153,17 +164,17 @@ "inputNode": { "id": -10, "bounding": [ - -80, - 425, + -560, + 480, 120, - 160 + 200 ] }, "outputNode": { "id": -20, "bounding": [ - 1490, - 415, + 1670, + 320, 120, 60 ] @@ -178,8 +189,8 @@ ], "label": "prompt", "pos": [ - 20, - 445 + -460, + 500 ] }, { @@ -190,8 +201,8 @@ 35 ], "pos": [ - 20, - 465 + -460, + 520 ] }, { @@ -202,44 +213,68 @@ 36 ], "pos": [ - 20, - 485 + -460, + 540 ] }, { - "id": "23087d15-8412-4fbd-b71e-9b6d7ef76de1", + "id": "f77677f7-6bf6-4c19-a71f-c4a553d5981e", + "name": "seed", + "type": "INT", + "linkIds": [ + 71 + ], + "pos": [ + -460, + 560 + ] + }, + { + "id": "ef9a9fb1-5983-4bc9-a60b-cf5aec48bff1", + "name": "steps", + "type": "INT", + "linkIds": [ + 72 + ], + "pos": [ + -460, + 580 + ] + }, + { + "id": "a20a1b30-785f-4a04-bb6d-3d61adab9764", "name": "unet_name", "type": "COMBO", "linkIds": [ - 38 + 73 ], "pos": [ - 20, - 505 + -460, + 600 ] }, { - "id": "0677f5c3-2a3f-43d4-98ac-a4c56d5efdc0", + "id": "4af8fc2b-4655-4086-8240-45f8cb38c6f6", "name": "clip_name", "type": "COMBO", "linkIds": [ - 39 + 74 ], "pos": [ - 20, - 525 + -460, + 620 ] }, { - "id": "c85c0445-2641-48b1-bbca-95057edf2fcf", + "id": "4d518693-2807-439c-9cb6-cffd23ccba2c", "name": "vae_name", "type": "COMBO", "linkIds": [ - 40 + 75 ], "pos": [ - 20, - 545 + -460, + 640 ] } ], @@ -253,8 +288,8 @@ ], "localized_name": "IMAGE", "pos": [ - 1510, - 435 + 1690, + 340 ] } ], @@ -264,15 +299,15 @@ "id": 30, "type": "CLIPLoader", "pos": [ - 109.99997264844609, - 329.99999029608756 + 30, + 420 ], "size": [ - 269.9869791666667, - 106 + 270, + 150 ], "flags": {}, - "order": 0, + "order": 7, "mode": 0, "inputs": [ { @@ -282,7 +317,7 @@ "widget": { "name": "clip_name" }, - "link": 39 + "link": 74 }, { "localized_name": "type", @@ -315,9 +350,9 @@ } ], "properties": { + "Node name for S&R": "CLIPLoader", "cnr_id": "comfy-core", "ver": "0.3.73", - "Node name for S&R": "CLIPLoader", "models": [ { "name": "qwen_3_4b.safetensors", @@ -343,15 +378,15 @@ "id": 29, "type": "VAELoader", "pos": [ - 109.99997264844609, - 479.9999847172637 + 30, + 650 ], "size": [ - 269.9869791666667, - 58 + 270, + 110 ], "flags": {}, - "order": 1, + "order": 6, "mode": 0, "inputs": [ { @@ -361,7 +396,7 @@ "widget": { "name": "vae_name" }, - "link": 40 + "link": 75 } ], "outputs": [ @@ -375,9 +410,9 @@ } ], "properties": { + "Node name for S&R": "VAELoader", "cnr_id": "comfy-core", "ver": "0.3.73", - "Node name for S&R": "VAELoader", "models": [ { "name": "ae.safetensors", @@ -401,12 +436,12 @@ "id": 33, "type": "ConditioningZeroOut", "pos": [ - 639.9999103333332, - 620.0000271257795 + 630, + 960 ], "size": [ - 204.134765625, - 26 + 230, + 80 ], "flags": {}, "order": 8, @@ -430,9 +465,9 @@ } ], "properties": { + "Node name for S&R": "ConditioningZeroOut", "cnr_id": "comfy-core", "ver": "0.3.73", - "Node name for S&R": "ConditioningZeroOut", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -440,22 +475,21 @@ "secondTabText": "Send Back", "secondTabOffset": 80, "secondTabWidth": 65 - }, - "widgets_values": [] + } }, { "id": 8, "type": "VAEDecode", "pos": [ - 1219.9999088104782, - 160.00009184959066 + 1320, + 230 ], "size": [ - 209.98697916666669, - 46 + 230, + 100 ], "flags": {}, - "order": 5, + "order": 1, "mode": 0, "inputs": [ { @@ -483,9 +517,9 @@ } ], "properties": { + "Node name for S&R": "VAEDecode", "cnr_id": "comfy-core", "ver": "0.3.64", - "Node name for S&R": "VAEDecode", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -493,22 +527,21 @@ "secondTabText": "Send Back", "secondTabOffset": 80, "secondTabWidth": 65 - }, - "widgets_values": [] + } }, { "id": 28, "type": "UNETLoader", "pos": [ - 109.99997264844609, - 200.0000502647102 + 30, + 230 ], "size": [ - 269.9869791666667, - 82 + 270, + 110 ], "flags": {}, - "order": 2, + "order": 5, "mode": 0, "inputs": [ { @@ -518,7 +551,7 @@ "widget": { "name": "unet_name" }, - "link": 38 + "link": 73 }, { "localized_name": "weight_dtype", @@ -541,9 +574,9 @@ } ], "properties": { + "Node name for S&R": "UNETLoader", "cnr_id": "comfy-core", "ver": "0.3.73", - "Node name for S&R": "UNETLoader", "models": [ { "name": "z_image_turbo_bf16.safetensors", @@ -568,15 +601,15 @@ "id": 27, "type": "CLIPTextEncode", "pos": [ - 429.99997828947767, - 200.0000502647102 + 400, + 230 ], "size": [ - 409.9869791666667, - 319.9869791666667 + 450, + 650 ], "flags": {}, - "order": 7, + "order": 4, "mode": 0, "inputs": [ { @@ -607,9 +640,9 @@ } ], "properties": { + "Node name for S&R": "CLIPTextEncode", "cnr_id": "comfy-core", "ver": "0.3.73", - "Node name for S&R": "CLIPTextEncode", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -626,15 +659,15 @@ "id": 13, "type": "EmptySD3LatentImage", "pos": [ - 109.99997264844609, - 629.9999791384399 + 40, + 890 ], "size": [ - 259.9869791666667, - 106 + 260, + 170 ], "flags": {}, - "order": 6, + "order": 3, "mode": 0, "inputs": [ { @@ -677,9 +710,9 @@ } ], "properties": { + "Node name for S&R": "EmptySD3LatentImage", "cnr_id": "comfy-core", "ver": "0.3.64", - "Node name for S&R": "EmptySD3LatentImage", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -694,19 +727,77 @@ 1 ] }, + { + "id": 11, + "type": "ModelSamplingAuraFlow", + "pos": [ + 950, + 230 + ], + "size": [ + 310, + 110 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 26 + }, + { + "localized_name": "shift", + "name": "shift", + "type": "FLOAT", + "widget": { + "name": "shift" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "slot_index": 0, + "links": [ + 13 + ] + } + ], + "properties": { + "Node name for S&R": "ModelSamplingAuraFlow", + "cnr_id": "comfy-core", + "ver": "0.3.64", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 3 + ] + }, { "id": 3, "type": "KSampler", "pos": [ - 879.9999615530063, - 269.9999774911694 + 950, + 400 ], "size": [ - 314.9869791666667, - 262 + 320, + 350 ], "flags": {}, - "order": 4, + "order": 0, "mode": 0, "inputs": [ { @@ -740,7 +831,7 @@ "widget": { "name": "seed" }, - "link": null + "link": 71 }, { "localized_name": "steps", @@ -749,7 +840,7 @@ "widget": { "name": "steps" }, - "link": null + "link": 72 }, { "localized_name": "cfg", @@ -800,9 +891,9 @@ } ], "properties": { + "Node name for S&R": "KSampler", "cnr_id": "comfy-core", "ver": "0.3.64", - "Node name for S&R": "KSampler", "enableTabs": false, "tabWidth": 65, "tabXOffset": 10, @@ -814,81 +905,23 @@ "widgets_values": [ 0, "randomize", - 4, + 8, 1, "res_multistep", "simple", 1 ] - }, - { - "id": 11, - "type": "ModelSamplingAuraFlow", - "pos": [ - 879.9999615530063, - 160.00009184959066 - ], - "size": [ - 309.9869791666667, - 58 - ], - "flags": {}, - "order": 3, - "mode": 0, - "inputs": [ - { - "localized_name": "model", - "name": "model", - "type": "MODEL", - "link": 26 - }, - { - "localized_name": "shift", - "name": "shift", - "type": "FLOAT", - "widget": { - "name": "shift" - }, - "link": null - } - ], - "outputs": [ - { - "localized_name": "MODEL", - "name": "MODEL", - "type": "MODEL", - "slot_index": 0, - "links": [ - 13 - ] - } - ], - "properties": { - "cnr_id": "comfy-core", - "ver": "0.3.64", - "Node name for S&R": "ModelSamplingAuraFlow", - "enableTabs": false, - "tabWidth": 65, - "tabXOffset": 10, - "hasSecondTab": false, - "secondTabText": "Send Back", - "secondTabOffset": 80, - "secondTabWidth": 65 - }, - "widgets_values": [ - 3 - ] } ], "groups": [ { "id": 2, - "title": "Image size", + "title": "Step2 - Image size", "bounding": [ - 100, - 560, - 290, - 200 + 10, + 820, + 320, + 280 ], "color": "#3f789e", "font_size": 24, @@ -896,12 +929,12 @@ }, { "id": 3, - "title": "Prompt", + "title": "Step3 - Prompt", "bounding": [ - 410, + 360, 130, - 450, - 540 + 530, + 970 ], "color": "#3f789e", "font_size": 24, @@ -909,12 +942,12 @@ }, { "id": 4, - "title": "Models", + "title": "Step1 - Load models", "bounding": [ - 100, + 0, 130, - 290, - 413.6 + 330, + 660 ], "color": "#3f789e", "font_size": 24, @@ -1027,25 +1060,41 @@ "type": "INT" }, { - "id": 38, + "id": 71, "origin_id": -10, "origin_slot": 3, + "target_id": 3, + "target_slot": 4, + "type": "INT" + }, + { + "id": 72, + "origin_id": -10, + "origin_slot": 4, + "target_id": 3, + "target_slot": 5, + "type": "INT" + }, + { + "id": 73, + "origin_id": -10, + "origin_slot": 5, "target_id": 28, "target_slot": 0, "type": "COMBO" }, { - "id": 39, + "id": 74, "origin_id": -10, - "origin_slot": 4, + "origin_slot": 6, "target_id": 30, "target_slot": 0, "type": "COMBO" }, { - "id": 40, + "id": 75, "origin_id": -10, - "origin_slot": 5, + "origin_slot": 7, "target_id": 29, "target_slot": 0, "type": "COMBO" @@ -1059,21 +1108,5 @@ } ] }, - "config": {}, - "extra": { - "frontendVersion": "1.37.10", - "workflowRendererVersion": "LG", - "VHS_latentpreview": false, - "VHS_latentpreviewrate": 0, - "VHS_MetadataImage": true, - "VHS_KeepIntermediate": true, - "ds": { - "scale": 0.8401370345180755, - "offset": [ - 940.0587067393087, - -830.7121087564725 - ] - } - }, - "version": 0.4 + "extra": {} } \ No newline at end of file diff --git a/blueprints/Text to Image.json b/blueprints/Text to Image.json new file mode 100644 index 000000000..ffe3682ff --- /dev/null +++ b/blueprints/Text to Image.json @@ -0,0 +1,1132 @@ +{ + "revision": 0, + "last_node_id": 71, + "last_link_id": 0, + "nodes": [ + { + "id": 71, + "type": "2d5985c9-deef-41ae-9c34-6353d3d7d1ef", + "pos": [ + 90, + 800 + ], + "size": [ + 400, + 80 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "label": "prompt", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": null + }, + { + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": null + }, + { + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": null + }, + { + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": null + }, + { + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": null + }, + { + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": null + }, + { + "name": "steps", + "type": "INT", + "widget": { + "name": "steps" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "links": [] + } + ], + "title": "Text to Image", + "properties": { + "proxyWidgets": [ + [ + "67", + "text" + ], + [ + "68", + "width" + ], + [ + "68", + "height" + ], + [ + "66", + "unet_name" + ], + [ + "62", + "clip_name" + ], + [ + "63", + "vae_name" + ], + [ + "70", + "steps" + ], + [ + "70", + "control_after_generate" + ] + ], + "cnr_id": "comfy-core", + "ver": "0.3.73", + "ue_properties": { + "widget_ue_connectable": { + "text": true + }, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [] + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "2d5985c9-deef-41ae-9c34-6353d3d7d1ef", + "version": 1, + "state": { + "lastGroupId": 4, + "lastNodeId": 71, + "lastLinkId": 70, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "Text to Image", + "inputNode": { + "id": -10, + "bounding": [ + -80, + 425, + 120, + 180 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + 1490, + 415, + 120, + 60 + ] + }, + "inputs": [ + { + "id": "fb178669-e742-4a53-8a69-7df59834dfd8", + "name": "text", + "type": "STRING", + "linkIds": [ + 34 + ], + "label": "prompt", + "pos": [ + 20, + 445 + ] + }, + { + "id": "dd780b3c-23e9-46ff-8469-156008f42e5a", + "name": "width", + "type": "INT", + "linkIds": [ + 35 + ], + "pos": [ + 20, + 465 + ] + }, + { + "id": "7b08d546-6bb0-4ef9-82e9-ffae5e1ee6bc", + "name": "height", + "type": "INT", + "linkIds": [ + 36 + ], + "pos": [ + 20, + 485 + ] + }, + { + "id": "8ed4eb73-a2bf-4766-8bf4-c5890b560596", + "name": "unet_name", + "type": "COMBO", + "linkIds": [ + 38 + ], + "pos": [ + 20, + 505 + ] + }, + { + "id": "f362d639-d412-4b5d-8490-1e9995dc5f82", + "name": "clip_name", + "type": "COMBO", + "linkIds": [ + 39 + ], + "pos": [ + 20, + 525 + ] + }, + { + "id": "ee25ac16-de63-4b74-bbbb-5b29fdc1efcf", + "name": "vae_name", + "type": "COMBO", + "linkIds": [ + 40 + ], + "pos": [ + 20, + 545 + ] + }, + { + "id": "51cbcd61-9218-4bcb-89ac-ecdfb1ef8892", + "name": "steps", + "type": "INT", + "linkIds": [ + 70 + ], + "pos": [ + 20, + 565 + ] + } + ], + "outputs": [ + { + "id": "1fa72a21-ce00-4952-814e-1f2ffbe87d1d", + "name": "IMAGE", + "type": "IMAGE", + "linkIds": [ + 16 + ], + "localized_name": "IMAGE", + "pos": [ + 1510, + 435 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 62, + "type": "CLIPLoader", + "pos": [ + 110, + 330 + ], + "size": [ + 270, + 110 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [ + { + "localized_name": "clip_name", + "name": "clip_name", + "type": "COMBO", + "widget": { + "name": "clip_name" + }, + "link": 39 + }, + { + "localized_name": "type", + "name": "type", + "type": "COMBO", + "widget": { + "name": "type" + }, + "link": null + }, + { + "localized_name": "device", + "name": "device", + "shape": 7, + "type": "COMBO", + "widget": { + "name": "device" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [ + 28 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.73", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "CLIPLoader", + "models": [ + { + "name": "qwen_3_4b.safetensors", + "url": "https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/text_encoders/qwen_3_4b.safetensors", + "directory": "text_encoders" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "qwen_3_4b.safetensors", + "lumina2", + "default" + ] + }, + { + "id": 63, + "type": "VAELoader", + "pos": [ + 110, + 480 + ], + "size": [ + 270, + 60 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "vae_name", + "name": "vae_name", + "type": "COMBO", + "widget": { + "name": "vae_name" + }, + "link": 40 + } + ], + "outputs": [ + { + "localized_name": "VAE", + "name": "VAE", + "type": "VAE", + "links": [ + 27 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.73", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "VAELoader", + "models": [ + { + "name": "ae.safetensors", + "url": "https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/vae/ae.safetensors", + "directory": "vae" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "ae.safetensors" + ] + }, + { + "id": 64, + "type": "ConditioningZeroOut", + "pos": [ + 640, + 620 + ], + "size": [ + 210, + 30 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "localized_name": "conditioning", + "name": "conditioning", + "type": "CONDITIONING", + "link": 32 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 33 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.73", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ConditioningZeroOut", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 65, + "type": "VAEDecode", + "pos": [ + 1220, + 160 + ], + "size": [ + 210, + 50 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "samples", + "name": "samples", + "type": "LATENT", + "link": 14 + }, + { + "localized_name": "vae", + "name": "vae", + "type": "VAE", + "link": 27 + } + ], + "outputs": [ + { + "localized_name": "IMAGE", + "name": "IMAGE", + "type": "IMAGE", + "slot_index": 0, + "links": [ + 16 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.64", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "VAEDecode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 66, + "type": "UNETLoader", + "pos": [ + 110, + 200 + ], + "size": [ + 270, + 90 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [ + { + "localized_name": "unet_name", + "name": "unet_name", + "type": "COMBO", + "widget": { + "name": "unet_name" + }, + "link": 38 + }, + { + "localized_name": "weight_dtype", + "name": "weight_dtype", + "type": "COMBO", + "widget": { + "name": "weight_dtype" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 26 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.73", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "UNETLoader", + "models": [ + { + "name": "z_image_turbo_bf16.safetensors", + "url": "https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/diffusion_models/z_image_turbo_bf16.safetensors", + "directory": "diffusion_models" + } + ], + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "z_image_turbo_bf16.safetensors", + "default" + ] + }, + { + "id": 67, + "type": "CLIPTextEncode", + "pos": [ + 430, + 200 + ], + "size": [ + 410, + 370 + ], + "flags": {}, + "order": 5, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 28 + }, + { + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": 34 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 30, + 32 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.73", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "CLIPTextEncode", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "" + ] + }, + { + "id": 68, + "type": "EmptySD3LatentImage", + "pos": [ + 110, + 630 + ], + "size": [ + 260, + 110 + ], + "flags": {}, + "order": 6, + "mode": 0, + "inputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "widget": { + "name": "width" + }, + "link": 35 + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "widget": { + "name": "height" + }, + "link": 36 + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "widget": { + "name": "batch_size" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "slot_index": 0, + "links": [ + 17 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.64", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "EmptySD3LatentImage", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 1024, + 1024, + 1 + ] + }, + { + "id": 69, + "type": "ModelSamplingAuraFlow", + "pos": [ + 880, + 160 + ], + "size": [ + 310, + 60 + ], + "flags": {}, + "order": 7, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 26 + }, + { + "localized_name": "shift", + "name": "shift", + "type": "FLOAT", + "widget": { + "name": "shift" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "slot_index": 0, + "links": [ + 13 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.64", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "ModelSamplingAuraFlow", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 3 + ] + }, + { + "id": 70, + "type": "KSampler", + "pos": [ + 880, + 270 + ], + "size": [ + 320, + 270 + ], + "flags": {}, + "order": 8, + "mode": 0, + "inputs": [ + { + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 13 + }, + { + "localized_name": "positive", + "name": "positive", + "type": "CONDITIONING", + "link": 30 + }, + { + "localized_name": "negative", + "name": "negative", + "type": "CONDITIONING", + "link": 33 + }, + { + "localized_name": "latent_image", + "name": "latent_image", + "type": "LATENT", + "link": 17 + }, + { + "localized_name": "seed", + "name": "seed", + "type": "INT", + "widget": { + "name": "seed" + }, + "link": null + }, + { + "localized_name": "steps", + "name": "steps", + "type": "INT", + "widget": { + "name": "steps" + }, + "link": 70 + }, + { + "localized_name": "cfg", + "name": "cfg", + "type": "FLOAT", + "widget": { + "name": "cfg" + }, + "link": null + }, + { + "localized_name": "sampler_name", + "name": "sampler_name", + "type": "COMBO", + "widget": { + "name": "sampler_name" + }, + "link": null + }, + { + "localized_name": "scheduler", + "name": "scheduler", + "type": "COMBO", + "widget": { + "name": "scheduler" + }, + "link": null + }, + { + "localized_name": "denoise", + "name": "denoise", + "type": "FLOAT", + "widget": { + "name": "denoise" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "LATENT", + "name": "LATENT", + "type": "LATENT", + "slot_index": 0, + "links": [ + 14 + ] + } + ], + "properties": { + "cnr_id": "comfy-core", + "ver": "0.3.64", + "ue_properties": { + "widget_ue_connectable": {}, + "version": "7.7", + "input_ue_unconnectable": {} + }, + "Node name for S&R": "KSampler", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 0, + "randomize", + 8, + 1, + "res_multistep", + "simple", + 1 + ] + } + ], + "groups": [ + { + "id": 2, + "title": "Step2 - Image size", + "bounding": [ + 100, + 560, + 290, + 200 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 3, + "title": "Step3 - Prompt", + "bounding": [ + 410, + 130, + 450, + 540 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + }, + { + "id": 4, + "title": "Step1 - Load models", + "bounding": [ + 100, + 130, + 290, + 413.6 + ], + "color": "#3f789e", + "font_size": 24, + "flags": {} + } + ], + "links": [ + { + "id": 32, + "origin_id": 67, + "origin_slot": 0, + "target_id": 64, + "target_slot": 0, + "type": "CONDITIONING" + }, + { + "id": 26, + "origin_id": 66, + "origin_slot": 0, + "target_id": 69, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 14, + "origin_id": 70, + "origin_slot": 0, + "target_id": 65, + "target_slot": 0, + "type": "LATENT" + }, + { + "id": 27, + "origin_id": 63, + "origin_slot": 0, + "target_id": 65, + "target_slot": 1, + "type": "VAE" + }, + { + "id": 13, + "origin_id": 69, + "origin_slot": 0, + "target_id": 70, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 30, + "origin_id": 67, + "origin_slot": 0, + "target_id": 70, + "target_slot": 1, + "type": "CONDITIONING" + }, + { + "id": 33, + "origin_id": 64, + "origin_slot": 0, + "target_id": 70, + "target_slot": 2, + "type": "CONDITIONING" + }, + { + "id": 17, + "origin_id": 68, + "origin_slot": 0, + "target_id": 70, + "target_slot": 3, + "type": "LATENT" + }, + { + "id": 28, + "origin_id": 62, + "origin_slot": 0, + "target_id": 67, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 16, + "origin_id": 65, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 34, + "origin_id": -10, + "origin_slot": 0, + "target_id": 67, + "target_slot": 1, + "type": "STRING" + }, + { + "id": 35, + "origin_id": -10, + "origin_slot": 1, + "target_id": 68, + "target_slot": 0, + "type": "INT" + }, + { + "id": 36, + "origin_id": -10, + "origin_slot": 2, + "target_id": 68, + "target_slot": 1, + "type": "INT" + }, + { + "id": 38, + "origin_id": -10, + "origin_slot": 3, + "target_id": 66, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 39, + "origin_id": -10, + "origin_slot": 4, + "target_id": 62, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 40, + "origin_id": -10, + "origin_slot": 5, + "target_id": 63, + "target_slot": 0, + "type": "COMBO" + }, + { + "id": 70, + "origin_id": -10, + "origin_slot": 6, + "target_id": 70, + "target_slot": 5, + "type": "INT" + } + ], + "extra": { + "workflowRendererVersion": "LG" + }, + "category": "Image generation and editing/Text to image", + "description": "Generates images from text prompts using Z-Image-Turbo defaults with Qwen3 text encoder and VAE." + } + ] + }, + "extra": {} +} diff --git a/blueprints/Video Segmentation (SAM3).json b/blueprints/Video Segmentation (SAM3).json new file mode 100644 index 000000000..4d9a13412 --- /dev/null +++ b/blueprints/Video Segmentation (SAM3).json @@ -0,0 +1,827 @@ +{ + "revision": 0, + "last_node_id": 130, + "last_link_id": 0, + "nodes": [ + { + "id": 130, + "type": "7937cf78-b52b-40a3-93b2-b4e2e5f98df1", + "pos": [ + -1210, + -2780 + ], + "size": [ + 300, + 370 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "name": "video", + "type": "VIDEO", + "link": null + }, + { + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": null + }, + { + "name": "bboxes", + "type": "BOUNDING_BOX", + "link": null + }, + { + "name": "positive_coords", + "type": "STRING", + "link": null + }, + { + "name": "negative_coords", + "type": "STRING", + "link": null + }, + { + "name": "threshold", + "type": "FLOAT", + "widget": { + "name": "threshold" + }, + "link": null + }, + { + "name": "refine_iterations", + "type": "INT", + "widget": { + "name": "refine_iterations" + }, + "link": null + }, + { + "name": "individual_masks", + "type": "BOOLEAN", + "widget": { + "name": "individual_masks" + }, + "link": null + }, + { + "name": "ckpt_name", + "type": "COMBO", + "widget": { + "name": "ckpt_name" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "masks", + "name": "masks", + "type": "MASK", + "links": [] + }, + { + "localized_name": "bboxes", + "name": "bboxes", + "type": "BOUNDING_BOX", + "links": [] + }, + { + "name": "audio", + "type": "AUDIO", + "links": null + }, + { + "name": "fps", + "type": "FLOAT", + "links": null + } + ], + "properties": { + "proxyWidgets": [ + [ + "125", + "text" + ], + [ + "126", + "threshold" + ], + [ + "126", + "refine_iterations" + ], + [ + "126", + "individual_masks" + ], + [ + "127", + "ckpt_name" + ] + ], + "cnr_id": "comfy-core", + "ver": "0.19.3", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [], + "title": "Video Segmentation (SAM3)" + } + ], + "links": [], + "version": 0.4, + "definitions": { + "subgraphs": [ + { + "id": "7937cf78-b52b-40a3-93b2-b4e2e5f98df1", + "version": 1, + "state": { + "lastGroupId": 0, + "lastNodeId": 130, + "lastLinkId": 299, + "lastRerouteId": 0 + }, + "revision": 0, + "config": {}, + "name": "Video Segmentation (SAM3)", + "inputNode": { + "id": -10, + "bounding": [ + -2260, + -3450, + 136.369140625, + 220 + ] + }, + "outputNode": { + "id": -20, + "bounding": [ + -1050, + -3510, + 120, + 120 + ] + }, + "inputs": [ + { + "id": "680ffd88-32fe-48be-88d6-91ea44d5eaee", + "name": "video", + "type": "VIDEO", + "linkIds": [ + 252 + ], + "pos": [ + -2143.630859375, + -3430 + ] + }, + { + "id": "ceaf249c-32d7-4624-8bf6-e590e347ed90", + "name": "text", + "type": "STRING", + "linkIds": [ + 254 + ], + "pos": [ + -2143.630859375, + -3410 + ] + }, + { + "id": "1ffbff36-da0c-4854-8cb4-88ad31e64f99", + "name": "bboxes", + "type": "BOUNDING_BOX", + "linkIds": [ + 255 + ], + "pos": [ + -2143.630859375, + -3390 + ] + }, + { + "id": "67b7f4c7-cec0-4e00-b154-23cc1abf880e", + "name": "positive_coords", + "type": "STRING", + "linkIds": [ + 256 + ], + "pos": [ + -2143.630859375, + -3370 + ] + }, + { + "id": "b090a498-2bde-46b9-9554-18501401d687", + "name": "negative_coords", + "type": "STRING", + "linkIds": [ + 257 + ], + "pos": [ + -2143.630859375, + -3350 + ] + }, + { + "id": "1a76dfcf-ce95-46af-bba5-c42160c683dd", + "name": "threshold", + "type": "FLOAT", + "linkIds": [ + 261 + ], + "pos": [ + -2143.630859375, + -3330 + ] + }, + { + "id": "999523fa-c476-4c53-80c3-0a2f554d18ab", + "name": "refine_iterations", + "type": "INT", + "linkIds": [ + 262 + ], + "pos": [ + -2143.630859375, + -3310 + ] + }, + { + "id": "d2371011-7fe5-4a39-b0c1-df2e0bbd6ece", + "name": "individual_masks", + "type": "BOOLEAN", + "linkIds": [ + 263 + ], + "pos": [ + -2143.630859375, + -3290 + ] + }, + { + "id": "675a8b37-17db-48d1-853c-2fe5d6a74582", + "name": "ckpt_name", + "type": "COMBO", + "linkIds": [ + 273 + ], + "pos": [ + -2143.630859375, + -3270 + ] + } + ], + "outputs": [ + { + "id": "ff50da09-1e59-4a58-9b7f-be1a00aa5913", + "name": "masks", + "type": "MASK", + "linkIds": [ + 231 + ], + "localized_name": "masks", + "pos": [ + -1030, + -3490 + ] + }, + { + "id": "8f622e40-8528-4078-b7d3-147e9f872194", + "name": "bboxes", + "type": "BOUNDING_BOX", + "linkIds": [ + 232 + ], + "localized_name": "bboxes", + "pos": [ + -1030, + -3470 + ] + }, + { + "id": "6c9924ec-f0fa-4509-83ea-8f97f5889bcc", + "name": "audio", + "type": "AUDIO", + "linkIds": [ + 259 + ], + "pos": [ + -1030, + -3450 + ] + }, + { + "id": "82c1cddc-ab11-44eb-9e2f-1a5c7ea5645b", + "name": "fps", + "type": "FLOAT", + "linkIds": [ + 260 + ], + "pos": [ + -1030, + -3430 + ] + } + ], + "widgets": [], + "nodes": [ + { + "id": 125, + "type": "CLIPTextEncode", + "pos": [ + -2010, + -3040 + ], + "size": [ + 400, + 200 + ], + "flags": {}, + "order": 1, + "mode": 0, + "inputs": [ + { + "localized_name": "clip", + "name": "clip", + "type": "CLIP", + "link": 240 + }, + { + "localized_name": "text", + "name": "text", + "type": "STRING", + "widget": { + "name": "text" + }, + "link": 254 + } + ], + "outputs": [ + { + "localized_name": "CONDITIONING", + "name": "CONDITIONING", + "type": "CONDITIONING", + "links": [ + 200 + ] + } + ], + "properties": { + "Node name for S&R": "CLIPTextEncode", + "cnr_id": "comfy-core", + "ver": "0.19.3", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + "" + ] + }, + { + "id": 126, + "type": "SAM3_Detect", + "pos": [ + -1520, + -3520 + ], + "size": [ + 270, + 290 + ], + "flags": {}, + "order": 2, + "mode": 0, + "inputs": [ + { + "label": "model", + "localized_name": "model", + "name": "model", + "type": "MODEL", + "link": 237 + }, + { + "label": "image", + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 253 + }, + { + "label": "conditioning", + "localized_name": "conditioning", + "name": "conditioning", + "shape": 7, + "type": "CONDITIONING", + "link": 200 + }, + { + "label": "bboxes", + "localized_name": "bboxes", + "name": "bboxes", + "shape": 7, + "type": "BOUNDING_BOX", + "link": 255 + }, + { + "label": "positive_coords", + "localized_name": "positive_coords", + "name": "positive_coords", + "shape": 7, + "type": "STRING", + "link": 256 + }, + { + "label": "negative_coords", + "localized_name": "negative_coords", + "name": "negative_coords", + "shape": 7, + "type": "STRING", + "link": 257 + }, + { + "localized_name": "threshold", + "name": "threshold", + "type": "FLOAT", + "widget": { + "name": "threshold" + }, + "link": 261 + }, + { + "localized_name": "refine_iterations", + "name": "refine_iterations", + "type": "INT", + "widget": { + "name": "refine_iterations" + }, + "link": 262 + }, + { + "localized_name": "individual_masks", + "name": "individual_masks", + "type": "BOOLEAN", + "widget": { + "name": "individual_masks" + }, + "link": 263 + } + ], + "outputs": [ + { + "localized_name": "masks", + "name": "masks", + "type": "MASK", + "links": [ + 231 + ] + }, + { + "localized_name": "bboxes", + "name": "bboxes", + "type": "BOUNDING_BOX", + "links": [ + 232 + ] + } + ], + "properties": { + "Node name for S&R": "SAM3_Detect", + "cnr_id": "comfy-core", + "ver": "0.19.3", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + }, + "widgets_values": [ + 0.5, + 2, + false + ] + }, + { + "id": 127, + "type": "CheckpointLoaderSimple", + "pos": [ + -1970, + -3310 + ], + "size": [ + 330, + 160 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "ckpt_name", + "name": "ckpt_name", + "type": "COMBO", + "widget": { + "name": "ckpt_name" + }, + "link": 273 + } + ], + "outputs": [ + { + "localized_name": "MODEL", + "name": "MODEL", + "type": "MODEL", + "links": [ + 237 + ] + }, + { + "localized_name": "CLIP", + "name": "CLIP", + "type": "CLIP", + "links": [ + 240 + ] + }, + { + "localized_name": "VAE", + "name": "VAE", + "type": "VAE", + "links": null + } + ], + "properties": { + "Node name for S&R": "CheckpointLoaderSimple", + "cnr_id": "comfy-core", + "ver": "0.19.3", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65, + "models": [ + { + "name": "sam3.1_multiplex_fp16.safetensors", + "url": "https://huggingface.co/Comfy-Org/sam3.1/resolve/main/checkpoints/sam3.1_multiplex_fp16.safetensors", + "directory": "checkpoints" + } + ] + }, + "widgets_values": [ + "sam3.1_multiplex_fp16.safetensors" + ] + }, + { + "id": 128, + "type": "GetVideoComponents", + "pos": [ + -1910, + -3540 + ], + "size": [ + 230, + 120 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [ + { + "localized_name": "video", + "name": "video", + "type": "VIDEO", + "link": 252 + } + ], + "outputs": [ + { + "localized_name": "images", + "name": "images", + "type": "IMAGE", + "links": [ + 253 + ] + }, + { + "localized_name": "audio", + "name": "audio", + "type": "AUDIO", + "links": [ + 259 + ] + }, + { + "localized_name": "fps", + "name": "fps", + "type": "FLOAT", + "links": [ + 260 + ] + } + ], + "properties": { + "Node name for S&R": "GetVideoComponents", + "cnr_id": "comfy-core", + "ver": "0.19.3", + "enableTabs": false, + "tabWidth": 65, + "tabXOffset": 10, + "hasSecondTab": false, + "secondTabText": "Send Back", + "secondTabOffset": 80, + "secondTabWidth": 65 + } + }, + { + "id": 129, + "type": "Note", + "pos": [ + -1980, + -2790 + ], + "size": [ + 370, + 250 + ], + "flags": {}, + "order": 0, + "mode": 0, + "inputs": [], + "outputs": [], + "title": "Note: Prompt format", + "properties": {}, + "widgets_values": [ + "Max tokens for this model is only 32, to separately prompt multiple subjects you can separate prompts with comma, and set the max amount of objects detected for each prompt with :N\n\nFor example above test prompt finds 2 cakes, one apron, 4 window panels" + ], + "color": "#432", + "bgcolor": "#653" + } + ], + "groups": [], + "links": [ + { + "id": 237, + "origin_id": 127, + "origin_slot": 0, + "target_id": 126, + "target_slot": 0, + "type": "MODEL" + }, + { + "id": 200, + "origin_id": 125, + "origin_slot": 0, + "target_id": 126, + "target_slot": 2, + "type": "CONDITIONING" + }, + { + "id": 240, + "origin_id": 127, + "origin_slot": 1, + "target_id": 125, + "target_slot": 0, + "type": "CLIP" + }, + { + "id": 231, + "origin_id": 126, + "origin_slot": 0, + "target_id": -20, + "target_slot": 0, + "type": "MASK" + }, + { + "id": 232, + "origin_id": 126, + "origin_slot": 1, + "target_id": -20, + "target_slot": 1, + "type": "BOUNDING_BOX" + }, + { + "id": 252, + "origin_id": -10, + "origin_slot": 0, + "target_id": 128, + "target_slot": 0, + "type": "VIDEO" + }, + { + "id": 253, + "origin_id": 128, + "origin_slot": 0, + "target_id": 126, + "target_slot": 1, + "type": "IMAGE" + }, + { + "id": 254, + "origin_id": -10, + "origin_slot": 1, + "target_id": 125, + "target_slot": 1, + "type": "STRING" + }, + { + "id": 255, + "origin_id": -10, + "origin_slot": 2, + "target_id": 126, + "target_slot": 3, + "type": "BOUNDING_BOX" + }, + { + "id": 256, + "origin_id": -10, + "origin_slot": 3, + "target_id": 126, + "target_slot": 4, + "type": "STRING" + }, + { + "id": 257, + "origin_id": -10, + "origin_slot": 4, + "target_id": 126, + "target_slot": 5, + "type": "STRING" + }, + { + "id": 259, + "origin_id": 128, + "origin_slot": 1, + "target_id": -20, + "target_slot": 2, + "type": "AUDIO" + }, + { + "id": 260, + "origin_id": 128, + "origin_slot": 2, + "target_id": -20, + "target_slot": 3, + "type": "FLOAT" + }, + { + "id": 261, + "origin_id": -10, + "origin_slot": 5, + "target_id": 126, + "target_slot": 6, + "type": "FLOAT" + }, + { + "id": 262, + "origin_id": -10, + "origin_slot": 6, + "target_id": 126, + "target_slot": 7, + "type": "INT" + }, + { + "id": 263, + "origin_id": -10, + "origin_slot": 7, + "target_id": 126, + "target_slot": 8, + "type": "BOOLEAN" + }, + { + "id": 273, + "origin_id": -10, + "origin_slot": 8, + "target_id": 127, + "target_slot": 0, + "type": "COMBO" + } + ], + "extra": {}, + "category": "Video Tools", + "description": "Segments video into temporally consistent masks using Meta SAM3 from text or interactive prompts." + } + ] + }, + "extra": {} +} diff --git a/blueprints/Video Stitch.json b/blueprints/Video Stitch.json index 6eb0f0bbf..2ac78b328 100644 --- a/blueprints/Video Stitch.json +++ b/blueprints/Video Stitch.json @@ -1,21 +1,21 @@ { "revision": 0, - "last_node_id": 84, + "last_node_id": 85, "last_link_id": 0, "nodes": [ { - "id": 84, - "type": "8e8aa94a-647e-436d-8440-8ee4691864de", + "id": 85, + "type": "637913e7-0206-46ba-8ded-70ae3a7c2e19", "pos": [ - -6100, - 2620 + -880, + -2260 ], "size": [ 290, 160 ], "flags": {}, - "order": 0, + "order": 2, "mode": 0, "inputs": [ { @@ -76,31 +76,26 @@ "properties": { "proxyWidgets": [ [ - "-1", + "79", "direction" ], [ - "-1", + "79", "match_image_size" ], [ - "-1", + "79", "spacing_width" ], [ - "-1", + "79", "spacing_color" ] ], "cnr_id": "comfy-core", "ver": "0.13.0" }, - "widgets_values": [ - "right", - true, - 0, - "white" - ], + "widgets_values": [], "title": "Video Stitch" } ], @@ -109,12 +104,12 @@ "definitions": { "subgraphs": [ { - "id": "8e8aa94a-647e-436d-8440-8ee4691864de", + "id": "637913e7-0206-46ba-8ded-70ae3a7c2e19", "version": 1, "state": { "lastGroupId": 1, - "lastNodeId": 84, - "lastLinkId": 262, + "lastNodeId": 97, + "lastLinkId": 282, "lastRerouteId": 0 }, "revision": 0, @@ -123,8 +118,8 @@ "inputNode": { "id": -10, "bounding": [ - -6580, - 2649, + -6810, + 2580, 143.55859375, 160 ] @@ -132,8 +127,8 @@ "outputNode": { "id": -20, "bounding": [ - -5720, - 2659, + -4770, + 2600, 120, 60 ] @@ -149,8 +144,8 @@ "localized_name": "video", "label": "Before Video", "pos": [ - -6456.44140625, - 2669 + -6686.44140625, + 2600 ] }, { @@ -163,8 +158,8 @@ "localized_name": "video_1", "label": "After Video", "pos": [ - -6456.44140625, - 2689 + -6686.44140625, + 2620 ] }, { @@ -175,8 +170,8 @@ 259 ], "pos": [ - -6456.44140625, - 2709 + -6686.44140625, + 2640 ] }, { @@ -187,8 +182,8 @@ 260 ], "pos": [ - -6456.44140625, - 2729 + -6686.44140625, + 2660 ] }, { @@ -199,8 +194,8 @@ 261 ], "pos": [ - -6456.44140625, - 2749 + -6686.44140625, + 2680 ] }, { @@ -211,8 +206,8 @@ 262 ], "pos": [ - -6456.44140625, - 2769 + -6686.44140625, + 2700 ] } ], @@ -226,8 +221,8 @@ ], "localized_name": "VIDEO", "pos": [ - -5700, - 2679 + -4750, + 2620 ] } ], @@ -238,11 +233,11 @@ "type": "GetVideoComponents", "pos": [ -6390, - 2560 + 2600 ], "size": [ - 193.530859375, - 66 + 230, + 120 ], "flags": {}, "order": 1, @@ -278,9 +273,9 @@ } ], "properties": { + "Node name for S&R": "GetVideoComponents", "cnr_id": "comfy-core", - "ver": "0.13.0", - "Node name for S&R": "GetVideoComponents" + "ver": "0.13.0" } }, { @@ -291,8 +286,8 @@ 2420 ], "size": [ - 193.530859375, - 66 + 230, + 120 ], "flags": {}, "order": 0, @@ -332,21 +327,254 @@ } ], "properties": { + "Node name for S&R": "GetVideoComponents", "cnr_id": "comfy-core", - "ver": "0.13.0", - "Node name for S&R": "GetVideoComponents" + "ver": "0.13.0" } }, + { + "id": 90, + "type": "GetImageSize", + "pos": [ + -6390, + 3030 + ], + "size": [ + 230, + 120 + ], + "flags": {}, + "order": 4, + "mode": 0, + "inputs": [ + { + "localized_name": "image", + "name": "image", + "type": "IMAGE", + "link": 266 + } + ], + "outputs": [ + { + "localized_name": "width", + "name": "width", + "type": "INT", + "links": [ + 274 + ] + }, + { + "localized_name": "height", + "name": "height", + "type": "INT", + "links": [ + 276 + ] + }, + { + "localized_name": "batch_size", + "name": "batch_size", + "type": "INT", + "links": null + } + ], + "properties": { + "Node name for S&R": "GetImageSize" + } + }, + { + "id": 80, + "type": "CreateVideo", + "pos": [ + -5190, + 2420 + ], + "size": [ + 270, + 130 + ], + "flags": {}, + "order": 3, + "mode": 0, + "inputs": [ + { + "localized_name": "images", + "name": "images", + "type": "IMAGE", + "link": 282 + }, + { + "localized_name": "audio", + "name": "audio", + "shape": 7, + "type": "AUDIO", + "link": 251 + }, + { + "localized_name": "fps", + "name": "fps", + "type": "FLOAT", + "widget": { + "name": "fps" + }, + "link": 252 + } + ], + "outputs": [ + { + "localized_name": "VIDEO", + "name": "VIDEO", + "type": "VIDEO", + "links": [ + 255 + ] + } + ], + "properties": { + "Node name for S&R": "CreateVideo", + "cnr_id": "comfy-core", + "ver": "0.13.0" + }, + "widgets_values": [ + 30 + ] + }, + { + "id": 95, + "type": "ComfyMathExpression", + "pos": [ + -6040, + 3020 + ], + "size": [ + 400, + 200 + ], + "flags": {}, + "order": 5, + "mode": 0, + "inputs": [ + { + "label": "a", + "localized_name": "values.a", + "name": "values.a", + "type": "FLOAT,INT", + "link": 274 + }, + { + "label": "b", + "localized_name": "values.b", + "name": "values.b", + "shape": 7, + "type": "FLOAT,INT", + "link": null + }, + { + "localized_name": "expression", + "name": "expression", + "type": "STRING", + "widget": { + "name": "expression" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "FLOAT", + "name": "FLOAT", + "type": "FLOAT", + "links": null + }, + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 279 + ] + } + ], + "properties": { + "Node name for S&R": "ComfyMathExpression" + }, + "widgets_values": [ + "a & ~1" + ] + }, + { + "id": 96, + "type": "ComfyMathExpression", + "pos": [ + -6040, + 3290 + ], + "size": [ + 400, + 200 + ], + "flags": {}, + "order": 6, + "mode": 0, + "inputs": [ + { + "label": "a", + "localized_name": "values.a", + "name": "values.a", + "type": "FLOAT,INT", + "link": 276 + }, + { + "label": "b", + "localized_name": "values.b", + "name": "values.b", + "shape": 7, + "type": "FLOAT,INT", + "link": null + }, + { + "localized_name": "expression", + "name": "expression", + "type": "STRING", + "widget": { + "name": "expression" + }, + "link": null + } + ], + "outputs": [ + { + "localized_name": "FLOAT", + "name": "FLOAT", + "type": "FLOAT", + "links": null + }, + { + "localized_name": "INT", + "name": "INT", + "type": "INT", + "links": [ + 280 + ] + } + ], + "properties": { + "Node name for S&R": "ComfyMathExpression" + }, + "widgets_values": [ + "a & ~1" + ] + }, { "id": 79, "type": "ImageStitch", "pos": [ -6390, - 2700 + 2780 ], "size": [ 270, - 150 + 160 ], "flags": {}, "order": 2, @@ -408,14 +636,15 @@ "name": "IMAGE", "type": "IMAGE", "links": [ - 250 + 266, + 281 ] } ], "properties": { + "Node name for S&R": "ImageStitch", "cnr_id": "comfy-core", - "ver": "0.13.0", - "Node name for S&R": "ImageStitch" + "ver": "0.13.0" }, "widgets_values": [ "right", @@ -425,60 +654,91 @@ ] }, { - "id": 80, - "type": "CreateVideo", + "id": 97, + "type": "ResizeImageMaskNode", "pos": [ - -6040, - 2610 + -5560, + 2790 ], "size": [ 270, - 78 + 160 ], "flags": {}, - "order": 3, + "order": 7, "mode": 0, "inputs": [ { - "localized_name": "images", - "name": "images", - "type": "IMAGE", - "link": 250 + "localized_name": "input", + "name": "input", + "type": "IMAGE,MASK", + "link": 281 }, { - "localized_name": "audio", - "name": "audio", - "shape": 7, - "type": "AUDIO", - "link": 251 - }, - { - "localized_name": "fps", - "name": "fps", - "type": "FLOAT", + "localized_name": "resize_type", + "name": "resize_type", + "type": "COMFY_DYNAMICCOMBO_V3", "widget": { - "name": "fps" + "name": "resize_type" }, - "link": 252 + "link": null + }, + { + "localized_name": "width", + "name": "resize_type.width", + "type": "INT", + "widget": { + "name": "resize_type.width" + }, + "link": 279 + }, + { + "localized_name": "height", + "name": "resize_type.height", + "type": "INT", + "widget": { + "name": "resize_type.height" + }, + "link": 280 + }, + { + "localized_name": "crop", + "name": "resize_type.crop", + "type": "COMBO", + "widget": { + "name": "resize_type.crop" + }, + "link": null + }, + { + "localized_name": "scale_method", + "name": "scale_method", + "type": "COMBO", + "widget": { + "name": "scale_method" + }, + "link": null } ], "outputs": [ { - "localized_name": "VIDEO", - "name": "VIDEO", - "type": "VIDEO", + "localized_name": "resized", + "name": "resized", + "type": "*", "links": [ - 255 + 282 ] } ], "properties": { - "cnr_id": "comfy-core", - "ver": "0.13.0", - "Node name for S&R": "CreateVideo" + "Node name for S&R": "ResizeImageMaskNode" }, "widgets_values": [ - 30 + "scale dimensions", + 512, + 512, + "center", + "area" ] } ], @@ -500,14 +760,6 @@ "target_slot": 1, "type": "IMAGE" }, - { - "id": 250, - "origin_id": 79, - "origin_slot": 0, - "target_id": 80, - "target_slot": 0, - "type": "IMAGE" - }, { "id": 251, "origin_id": 77, @@ -579,6 +831,62 @@ "target_id": 79, "target_slot": 5, "type": "COMBO" + }, + { + "id": 266, + "origin_id": 79, + "origin_slot": 0, + "target_id": 90, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 274, + "origin_id": 90, + "origin_slot": 0, + "target_id": 95, + "target_slot": 0, + "type": "INT" + }, + { + "id": 276, + "origin_id": 90, + "origin_slot": 1, + "target_id": 96, + "target_slot": 0, + "type": "INT" + }, + { + "id": 279, + "origin_id": 95, + "origin_slot": 1, + "target_id": 97, + "target_slot": 2, + "type": "INT" + }, + { + "id": 280, + "origin_id": 96, + "origin_slot": 1, + "target_id": 97, + "target_slot": 3, + "type": "INT" + }, + { + "id": 281, + "origin_id": 79, + "origin_slot": 0, + "target_id": 97, + "target_slot": 0, + "type": "IMAGE" + }, + { + "id": 282, + "origin_id": 97, + "origin_slot": 0, + "target_id": 80, + "target_slot": 0, + "type": "IMAGE" } ], "extra": { @@ -588,5 +896,6 @@ "description": "Stitches multiple video clips into a single sequential video file." } ] - } + }, + "extra": {} } \ No newline at end of file diff --git a/comfy/k_diffusion/sampling.py b/comfy/k_diffusion/sampling.py index c53ac4b2b..11db46d94 100644 --- a/comfy/k_diffusion/sampling.py +++ b/comfy/k_diffusion/sampling.py @@ -242,6 +242,7 @@ def sample_euler_ancestral_RF(model, x, sigmas, extra_args=None, callback=None, extra_args = {} if extra_args is None else extra_args seed = extra_args.get("seed", None) noise_sampler = default_noise_sampler(x, seed=seed) if noise_sampler is None else noise_sampler + s_noise = s_noise * getattr(model.inner_model.model_patcher.get_model_object('model_sampling'), "noise_scale", 1.0) s_in = x.new_ones([x.shape[0]]) for i in trange(len(sigmas) - 1, disable=disable): denoised = model(x, sigmas[i] * s_in, **extra_args) @@ -373,6 +374,7 @@ def sample_dpm_2_ancestral_RF(model, x, sigmas, extra_args=None, callback=None, extra_args = {} if extra_args is None else extra_args seed = extra_args.get("seed", None) noise_sampler = default_noise_sampler(x, seed=seed) if noise_sampler is None else noise_sampler + s_noise = s_noise * getattr(model.inner_model.model_patcher.get_model_object('model_sampling'), "noise_scale", 1.0) s_in = x.new_ones([x.shape[0]]) for i in trange(len(sigmas) - 1, disable=disable): denoised = model(x, sigmas[i] * s_in, **extra_args) @@ -686,6 +688,7 @@ def sample_dpmpp_2s_ancestral_RF(model, x, sigmas, extra_args=None, callback=Non extra_args = {} if extra_args is None else extra_args seed = extra_args.get("seed", None) noise_sampler = default_noise_sampler(x, seed=seed) if noise_sampler is None else noise_sampler + s_noise = s_noise * getattr(model.inner_model.model_patcher.get_model_object('model_sampling'), "noise_scale", 1.0) s_in = x.new_ones([x.shape[0]]) sigma_fn = lambda lbda: (lbda.exp() + 1) ** -1 lambda_fn = lambda sigma: ((1-sigma)/sigma).log() @@ -747,6 +750,7 @@ def sample_dpmpp_sde(model, x, sigmas, extra_args=None, callback=None, disable=N sigma_fn = partial(half_log_snr_to_sigma, model_sampling=model_sampling) lambda_fn = partial(sigma_to_half_log_snr, model_sampling=model_sampling) sigmas = offset_first_sigma_for_snr(sigmas, model_sampling) + s_noise = s_noise * getattr(model_sampling, "noise_scale", 1.0) for i in trange(len(sigmas) - 1, disable=disable): denoised = model(x, sigmas[i] * s_in, **extra_args) @@ -832,6 +836,7 @@ def sample_dpmpp_2m_sde(model, x, sigmas, extra_args=None, callback=None, disabl model_sampling = model.inner_model.model_patcher.get_model_object('model_sampling') lambda_fn = partial(sigma_to_half_log_snr, model_sampling=model_sampling) sigmas = offset_first_sigma_for_snr(sigmas, model_sampling) + s_noise = s_noise * getattr(model_sampling, "noise_scale", 1.0) old_denoised = None h, h_last = None, None @@ -889,6 +894,7 @@ def sample_dpmpp_3m_sde(model, x, sigmas, extra_args=None, callback=None, disabl model_sampling = model.inner_model.model_patcher.get_model_object('model_sampling') lambda_fn = partial(sigma_to_half_log_snr, model_sampling=model_sampling) sigmas = offset_first_sigma_for_snr(sigmas, model_sampling) + s_noise = s_noise * getattr(model_sampling, "noise_scale", 1.0) denoised_1, denoised_2 = None, None h, h_1, h_2 = None, None, None @@ -1006,23 +1012,39 @@ def sample_ddpm(model, x, sigmas, extra_args=None, callback=None, disable=None, return generic_step_sampler(model, x, sigmas, extra_args, callback, disable, noise_sampler, DDPMSampler_step) @torch.no_grad() -def sample_lcm(model, x, sigmas, extra_args=None, callback=None, disable=None, noise_sampler=None): +def sample_lcm(model, x, sigmas, extra_args=None, callback=None, disable=None, noise_sampler=None, s_noise=1.0, s_noise_end=None, noise_clip_std=0.0): + + # s_noise / s_noise_end: per-step noise multiplier, linearly interpolated across steps + # noise_clip_std: clamp injected noise to +/- N stddevs (0 disables). + extra_args = {} if extra_args is None else extra_args seed = extra_args.get("seed", None) noise_sampler = default_noise_sampler(x, seed=seed) if noise_sampler is None else noise_sampler s_in = x.new_ones([x.shape[0]]) - for i in trange(len(sigmas) - 1, disable=disable): + n_steps = max(1, len(sigmas) - 1) + model_sampling = model.inner_model.model_patcher.get_model_object('model_sampling') + + s_start = float(s_noise) + s_end = s_start if s_noise_end is None else float(s_noise_end) + for i in trange(n_steps, disable=disable): denoised = model(x, sigmas[i] * s_in, **extra_args) if callback is not None: callback({'x': x, 'i': i, 'sigma': sigmas[i], 'sigma_hat': sigmas[i], 'denoised': denoised}) x = denoised if sigmas[i + 1] > 0: - x = model.inner_model.inner_model.model_sampling.noise_scaling(sigmas[i + 1], noise_sampler(sigmas[i], sigmas[i + 1]), x) + noise = noise_sampler(sigmas[i], sigmas[i + 1]) + if noise_clip_std > 0: + clip_val = noise_clip_std * noise.std() + noise = noise.clamp(min=-clip_val, max=clip_val) + t = (i / (n_steps - 1)) if n_steps > 1 else 0.0 + s_noise_i = s_start + (s_end - s_start) * t + if s_noise_i != 1.0: + noise = noise * s_noise_i + x = model_sampling.noise_scaling(sigmas[i + 1], noise, x) return x - @torch.no_grad() def sample_heunpp2(model, x, sigmas, extra_args=None, callback=None, disable=None, s_churn=0., s_tmin=0., s_tmax=float('inf'), s_noise=1.): # From MIT licensed: https://github.com/Carzit/sd-webui-samplers-scheduler/ @@ -1249,6 +1271,7 @@ def sample_euler_ancestral_cfg_pp(model, x, sigmas, extra_args=None, callback=No model_sampling = model.inner_model.model_patcher.get_model_object("model_sampling") lambda_fn = partial(sigma_to_half_log_snr, model_sampling=model_sampling) + s_noise = s_noise * getattr(model_sampling, "noise_scale", 1.0) uncond_denoised = None @@ -1296,6 +1319,7 @@ def sample_dpmpp_2s_ancestral_cfg_pp(model, x, sigmas, extra_args=None, callback extra_args = {} if extra_args is None else extra_args seed = extra_args.get("seed", None) noise_sampler = default_noise_sampler(x, seed=seed) if noise_sampler is None else noise_sampler + s_noise = s_noise * getattr(model.inner_model.model_patcher.get_model_object('model_sampling'), "noise_scale", 1.0) temp = [0] def post_cfg_function(args): @@ -1371,6 +1395,7 @@ def res_multistep(model, x, sigmas, extra_args=None, callback=None, disable=None extra_args = {} if extra_args is None else extra_args seed = extra_args.get("seed", None) noise_sampler = default_noise_sampler(x, seed=seed) if noise_sampler is None else noise_sampler + s_noise = s_noise * getattr(model.inner_model.model_patcher.get_model_object('model_sampling'), "noise_scale", 1.0) s_in = x.new_ones([x.shape[0]]) sigma_fn = lambda t: t.neg().exp() t_fn = lambda sigma: sigma.log().neg() @@ -1504,6 +1529,7 @@ def sample_er_sde(model, x, sigmas, extra_args=None, callback=None, disable=None extra_args = {} if extra_args is None else extra_args seed = extra_args.get("seed", None) noise_sampler = default_noise_sampler(x, seed=seed) if noise_sampler is None else noise_sampler + s_noise = s_noise * getattr(model.inner_model.model_patcher.get_model_object('model_sampling'), "noise_scale", 1.0) s_in = x.new_ones([x.shape[0]]) def default_er_sde_noise_scaler(x): @@ -1574,9 +1600,10 @@ def sample_seeds_2(model, x, sigmas, extra_args=None, callback=None, disable=Non seed = extra_args.get("seed", None) noise_sampler = default_noise_sampler(x, seed=seed) if noise_sampler is None else noise_sampler s_in = x.new_ones([x.shape[0]]) - inject_noise = eta > 0 and s_noise > 0 model_sampling = model.inner_model.model_patcher.get_model_object('model_sampling') + s_noise = s_noise * getattr(model_sampling, "noise_scale", 1.0) + inject_noise = eta > 0 and s_noise > 0 sigma_fn = partial(half_log_snr_to_sigma, model_sampling=model_sampling) lambda_fn = partial(sigma_to_half_log_snr, model_sampling=model_sampling) sigmas = offset_first_sigma_for_snr(sigmas, model_sampling) @@ -1645,9 +1672,10 @@ def sample_seeds_3(model, x, sigmas, extra_args=None, callback=None, disable=Non seed = extra_args.get("seed", None) noise_sampler = default_noise_sampler(x, seed=seed) if noise_sampler is None else noise_sampler s_in = x.new_ones([x.shape[0]]) - inject_noise = eta > 0 and s_noise > 0 model_sampling = model.inner_model.model_patcher.get_model_object('model_sampling') + s_noise = s_noise * getattr(model_sampling, "noise_scale", 1.0) + inject_noise = eta > 0 and s_noise > 0 sigma_fn = partial(half_log_snr_to_sigma, model_sampling=model_sampling) lambda_fn = partial(sigma_to_half_log_snr, model_sampling=model_sampling) sigmas = offset_first_sigma_for_snr(sigmas, model_sampling) @@ -1713,6 +1741,7 @@ def sample_sa_solver(model, x, sigmas, extra_args=None, callback=None, disable=F s_in = x.new_ones([x.shape[0]]) model_sampling = model.inner_model.model_patcher.get_model_object("model_sampling") + s_noise = s_noise * getattr(model_sampling, "noise_scale", 1.0) sigmas = offset_first_sigma_for_snr(sigmas, model_sampling) lambdas = sigma_to_half_log_snr(sigmas, model_sampling=model_sampling) diff --git a/comfy/latent_formats.py b/comfy/latent_formats.py index 91bebed3d..d527eec4a 100644 --- a/comfy/latent_formats.py +++ b/comfy/latent_formats.py @@ -792,6 +792,13 @@ class ZImagePixelSpace(ChromaRadiance): """ pass + +class HiDreamO1Pixel(ChromaRadiance): + """Pixel-space latent format for HiDream-O1. + No VAE — model patches/unpatches raw RGB internally with patch_size=32. + """ + pass + class CogVideoX(LatentFormat): """Latent format for CogVideoX-2b (THUDM/CogVideoX-2b). diff --git a/comfy/ldm/hidream_o1/attention.py b/comfy/ldm/hidream_o1/attention.py new file mode 100644 index 000000000..1b68f1771 --- /dev/null +++ b/comfy/ldm/hidream_o1/attention.py @@ -0,0 +1,41 @@ +"""HiDream-O1 two-pass attention: tokens [0, ar_len) are causal, [ar_len, T) +attend full K/V. Splitting Q at the boundary avoids the (B, 1, T, T) additive +mask the general-purpose path would build (~500 MB at T~16K) and lets the +gen half hit the user's preferred backend via optimized_attention. +""" + +import torch + +import comfy.ops +from comfy.ldm.modules.attention import optimized_attention + + +def make_two_pass_attention(ar_len: int, transformer_options=None): + """Build a two-pass attention callable. AR pass uses SDPA-causal directly, gen pass routes through optimized_attention. + The AR pass goes through SDPA directand bypasses wrappers, it is only ~1% of T at typical edit sizes. + """ + + def two_pass_attention(q, k, v, heads, **kwargs): + B, H, T, D = q.shape + + if T < k.shape[2]: # KV-cache hot path: Q is shorter than K/V (cached AR prefix is in K/V only), all fresh Q positions are in the gen region, single full-attention call + out = optimized_attention(q, k, v, heads, mask=None, skip_reshape=True, skip_output_reshape=True, transformer_options=transformer_options) + elif ar_len >= T: + out = comfy.ops.scaled_dot_product_attention(q, k, v, attn_mask=None, dropout_p=0.0, is_causal=True) + elif ar_len <= 0: + out = optimized_attention(q, k, v, heads, mask=None, skip_reshape=True, skip_output_reshape=True, transformer_options=transformer_options) + else: + out_ar = comfy.ops.scaled_dot_product_attention( + q[:, :, :ar_len], k[:, :, :ar_len], v[:, :, :ar_len], + attn_mask=None, dropout_p=0.0, is_causal=True, + ) + out_gen = optimized_attention( + q[:, :, ar_len:], k, v, heads, + mask=None, skip_reshape=True, skip_output_reshape=True, + transformer_options=transformer_options, + ) + out = torch.cat([out_ar, out_gen], dim=2) + + return out.transpose(1, 2).reshape(B, T, H * D) + + return two_pass_attention diff --git a/comfy/ldm/hidream_o1/conditioning.py b/comfy/ldm/hidream_o1/conditioning.py new file mode 100644 index 000000000..7496f0035 --- /dev/null +++ b/comfy/ldm/hidream_o1/conditioning.py @@ -0,0 +1,230 @@ +"""HiDream-O1 conditioning prep — ref-image dual path + extra_conds assembly. + +Each ref image goes through two paths: a 32x32 patchified stream concatenated +to the noised target, and a Qwen3-VL ViT path producing tokens that scatter +into input_ids at <|image_pad|> positions. +""" + +from typing import List + +import torch + +import comfy.utils +from comfy.text_encoders.qwen_vl import process_qwen2vl_images + +from .utils import (PATCH_SIZE, calculate_dimensions, cond_image_size, ref_max_size, resize_tensor) + +# Qwen3-VL ViT preprocessing constants (preprocessor_config.json). +VIT_PATCH = 16 +VIT_MERGE = 2 +VIT_IMAGE_MEAN = [0.5, 0.5, 0.5] +VIT_IMAGE_STD = [0.5, 0.5, 0.5] + + +def prepare_ref_images( + ref_images: List[torch.Tensor], + target_h: int, + target_w: int, + device: torch.device, + dtype: torch.dtype, +): + """Build the dual-path tensors for K reference images at (target_h, target_w). + + Returns None for K=0, else a dict with ref_patches, ref_pixel_values, + ref_image_grid_thw, per_ref_vit_tokens, per_ref_patch_grids. + """ + K = len(ref_images) + if K == 0: + return None + max_size = ref_max_size(max(target_h, target_w), K) + cis = cond_image_size(K) + + refs_t = [img[0].clamp(0, 1).permute(2, 0, 1).unsqueeze(0).contiguous().float() for img in ref_images] + refs_t = [resize_tensor(t, max_size, PATCH_SIZE) for t in refs_t] + + # 32-patch path. + ref_patches_per = [] + per_ref_patch_grids = [] + for t in refs_t: + t_norm = (t.squeeze(0) - 0.5) / 0.5 # (3, H, W) in [-1, 1] + h_p, w_p = t_norm.shape[-2] // PATCH_SIZE, t_norm.shape[-1] // PATCH_SIZE + per_ref_patch_grids.append((h_p, w_p)) + patches = ( + t_norm.reshape(3, h_p, PATCH_SIZE, w_p, PATCH_SIZE) + .permute(1, 3, 0, 2, 4) + .reshape(h_p * w_p, 3 * PATCH_SIZE * PATCH_SIZE) + ) + ref_patches_per.append(patches) + ref_patches = torch.cat(ref_patches_per, dim=0).unsqueeze(0).to(device=device, dtype=dtype) + + # ViT path. + refs_vlm_t = [] + for t in refs_t: + _, _, h, w = t.shape + cond_w, cond_h = calculate_dimensions(cis, w / h) + cond_w = max(cond_w, VIT_PATCH * VIT_MERGE) + cond_h = max(cond_h, VIT_PATCH * VIT_MERGE) + refs_vlm_t.append(comfy.utils.common_upscale(t, cond_w, cond_h, "lanczos", "disabled")) + + pv_list, grid_list, per_ref_vit_tokens = [], [], [] + for t_v in refs_vlm_t: + pv, grid_thw = process_qwen2vl_images( + t_v.permute(0, 2, 3, 1), + min_pixels=0, max_pixels=10**12, + patch_size=VIT_PATCH, merge_size=VIT_MERGE, + image_mean=VIT_IMAGE_MEAN, image_std=VIT_IMAGE_STD, + ) + grid_thw = grid_thw[0] + pv_list.append(pv.to(device=device, dtype=dtype)) + grid_list.append(grid_thw.to(device=device)) + # Post-merge token count = number of <|image_pad|> tokens this image expands to in input_ids. + gh, gw = int(grid_thw[1].item()), int(grid_thw[2].item()) + per_ref_vit_tokens.append((gh // VIT_MERGE) * (gw // VIT_MERGE)) + + return { + "ref_patches": ref_patches, + "ref_pixel_values": torch.cat(pv_list, dim=0), + "ref_image_grid_thw": torch.stack(grid_list, dim=0), + "per_ref_vit_tokens": per_ref_vit_tokens, + "per_ref_patch_grids": per_ref_patch_grids, + } + + +def build_ref_input_ids( + text_input_ids: torch.Tensor, + per_ref_vit_tokens: List[int], + image_token_id: int, + vision_start_id: int, + vision_end_id: int, +): + """Splice [vision_start, image_pad*N, vision_end] blocks into input_ids + after the [im_start, user, \\n] prefix (matches original chat template). + """ + ids = text_input_ids[0].tolist() + inserted = [] + for n_pad in per_ref_vit_tokens: + inserted.extend([vision_start_id] + [image_token_id] * n_pad + [vision_end_id]) + new_ids = ids[:3] + inserted + ids[3:] # 3 = len([im_start, user, \n]) + return torch.tensor([new_ids], dtype=text_input_ids.dtype, device=text_input_ids.device) + + +def build_extra_conds( + text_input_ids: torch.Tensor, + noise: torch.Tensor, + ref_images: List[torch.Tensor] = None, + target_patch_size: int = 32, +): + """Assemble all conditioning tensors for HiDreamO1Transformer.forward: + input_ids (with ref-vision tokens spliced in for the edit/IP path), + position_ids (MRoPE), token_types, vinput_mask, plus the ref + dual-path tensors when refs are provided. + """ + from .utils import get_rope_index_fix_point + from comfy.text_encoders.hidream_o1 import ( + IMAGE_TOKEN_ID, VISION_START_ID, VISION_END_ID, + ) + + if text_input_ids.dim() == 1: + text_input_ids = text_input_ids.unsqueeze(0) + text_input_ids = text_input_ids.long().to(noise.device) + B = noise.shape[0] + if text_input_ids.shape[0] == 1 and B > 1: + text_input_ids = text_input_ids.expand(B, -1) + + H, W = noise.shape[-2], noise.shape[-1] + h_p, w_p = H // target_patch_size, W // target_patch_size + image_len = h_p * w_p + image_grid_thw_tgt = torch.tensor( + [[1, h_p, w_p]], dtype=torch.long, device=text_input_ids.device, + ) + + out = {} + if ref_images: + ref = prepare_ref_images(ref_images, H, W, device=noise.device, dtype=noise.dtype) + text_input_ids = build_ref_input_ids( + text_input_ids, ref["per_ref_vit_tokens"], + IMAGE_TOKEN_ID, VISION_START_ID, VISION_END_ID, + ) + new_txt_len = text_input_ids.shape[1] + + # Each ref's patchified stream gets a [vision_start, image_pad*N-1] + # block in the position-id stream after the noised target. + ref_grid_lengths = [hp * wp for (hp, wp) in ref["per_ref_patch_grids"]] + tgt_vision = torch.full((1, image_len), IMAGE_TOKEN_ID, + dtype=text_input_ids.dtype, device=text_input_ids.device) + tgt_vision[:, 0] = VISION_START_ID + ref_vision_blocks = [] + for rl in ref_grid_lengths: + blk = torch.full((1, rl), IMAGE_TOKEN_ID, + dtype=text_input_ids.dtype, device=text_input_ids.device) + blk[:, 0] = VISION_START_ID + ref_vision_blocks.append(blk) + ref_vision_cat = torch.cat([tgt_vision] + ref_vision_blocks, dim=1) + input_ids_pad = torch.cat([text_input_ids, ref_vision_cat], dim=-1) + total_ref_patches_len = sum(ref_grid_lengths) + total_len = new_txt_len + image_len + total_ref_patches_len + + # K (ViT, post-merge) + 1 (target) + K (ref-patches) image grids. + K = len(ref_images) + igthw_cond = ref["ref_image_grid_thw"].clone() + igthw_cond[:, 1] //= 2 + igthw_cond[:, 2] //= 2 + image_grid_thw_ref = torch.tensor( + [[1, hp, wp] for (hp, wp) in ref["per_ref_patch_grids"]], + dtype=torch.long, device=text_input_ids.device, + ) + igthw_all = torch.cat([ + igthw_cond.to(text_input_ids.device), + image_grid_thw_tgt, + image_grid_thw_ref, + ], dim=0) + position_ids, _ = get_rope_index_fix_point( + spatial_merge_size=1, + image_token_id=IMAGE_TOKEN_ID, + vision_start_token_id=VISION_START_ID, + input_ids=input_ids_pad, image_grid_thw=igthw_all, + attention_mask=None, + skip_vision_start_token=[0] * K + [1] + [1] * K, + fix_point=4096, + ) + + # tms + target_image + ref_patches are all gen. + tms_pos = new_txt_len - 1 + ar_len = tms_pos + token_types = torch.zeros(B, total_len, dtype=torch.long, device=noise.device) + token_types[:, tms_pos:] = 1 + vinput_mask = torch.zeros(B, total_len, dtype=torch.bool, device=noise.device) + vinput_mask[:, new_txt_len:] = True + + # Leading batch dim sidesteps CONDRegular.process_cond's repeat_to_batch_size truncation + out["ref_pixel_values"] = ref["ref_pixel_values"].unsqueeze(0) + out["ref_image_grid_thw"] = ref["ref_image_grid_thw"].unsqueeze(0) + out["ref_patches"] = ref["ref_patches"] + else: + # T2I: text + noised target only, vision_start replaces the first image token + txt_len = text_input_ids.shape[1] + total_len = txt_len + image_len + vision_tokens = torch.full((B, image_len), IMAGE_TOKEN_ID, + dtype=text_input_ids.dtype, device=text_input_ids.device) + vision_tokens[:, 0] = VISION_START_ID + input_ids_pad = torch.cat([text_input_ids, vision_tokens], dim=-1) + position_ids, _ = get_rope_index_fix_point( + spatial_merge_size=1, + image_token_id=IMAGE_TOKEN_ID, + vision_start_token_id=VISION_START_ID, + input_ids=input_ids_pad, image_grid_thw=image_grid_thw_tgt, + attention_mask=None, + skip_vision_start_token=[1], + ) + ar_len = txt_len - 1 + token_types = torch.zeros(B, total_len, dtype=torch.long, device=noise.device) + token_types[:, ar_len:] = 1 + vinput_mask = torch.zeros(B, total_len, dtype=torch.bool, device=noise.device) + vinput_mask[:, txt_len:] = True + + out["input_ids"] = text_input_ids + out["position_ids"] = position_ids[:, 0].unsqueeze(0) # Collapse position_ids batch and add a leading dim so CONDRegular's batch-resize doesn't truncate the 3-axis MRoPE dim + out["token_types"] = token_types + out["vinput_mask"] = vinput_mask + out["ar_len"] = ar_len + return out diff --git a/comfy/ldm/hidream_o1/model.py b/comfy/ldm/hidream_o1/model.py new file mode 100644 index 000000000..a223e706f --- /dev/null +++ b/comfy/ldm/hidream_o1/model.py @@ -0,0 +1,306 @@ +"""HiDream-O1-Image transformer. + +Pixel-space DiT built on Qwen3-VL: the vision tower (Qwen35VisionModel) +encodes ref images, the Qwen3-VL-8B decoder (Llama2_ with interleaved MRoPE) +processes a unified text+image sequence, and 32x32 patch embed/unembed +shims map raw RGB in and out of LLM hidden space. The Qwen3-VL deepstack +mergers go unused — their weights are dropped at load. +""" + +from dataclasses import dataclass, field +from typing import List, Optional + +import einops +import torch +import torch.nn as nn + +import comfy.patcher_extension +from comfy.ldm.modules.diffusionmodules.mmdit import TimestepEmbedder +from comfy.text_encoders.llama import Llama2_ +from comfy.text_encoders.qwen35 import Qwen35VisionModel + +from .attention import make_two_pass_attention + + +IMAGE_TOKEN_ID = 151655 # Qwen3-VL <|image_pad|> +TMS_TOKEN_ID = 151673 # HiDream-O1 <|tms_token|> +PATCH_SIZE = 32 + + +@dataclass +class HiDreamO1TextConfig: + """Qwen3-VL-8B text-decoder dims (matches public Qwen3-VL-8B-Instruct).""" + vocab_size: int = 151936 + hidden_size: int = 4096 + intermediate_size: int = 12288 + num_hidden_layers: int = 36 + num_attention_heads: int = 32 + num_key_value_heads: int = 8 + head_dim: int = 128 + max_position_embeddings: int = 128000 + rms_norm_eps: float = 1e-6 + rope_theta: float = 5000000.0 + rope_scale: Optional[float] = None + rope_dims: List[int] = field(default_factory=lambda: [24, 20, 20]) + interleaved_mrope: bool = True + transformer_type: str = "llama" + rms_norm_add: bool = False + mlp_activation: str = "silu" + qkv_bias: bool = False + q_norm: str = "gemma3" + k_norm: str = "gemma3" + final_norm: bool = True + lm_head: bool = False + stop_tokens: List[int] = field(default_factory=lambda: [151643, 151645]) + + +QWEN3VL_VISION_DEFAULTS = dict( + hidden_size=1152, + num_heads=16, + intermediate_size=4304, + depth=27, + patch_size=16, + temporal_patch_size=2, + in_channels=3, + spatial_merge_size=2, + num_position_embeddings=2304, + deepstack_visual_indexes=(8, 16, 24), + out_hidden_size=4096, # final merger projects directly into LLM hidden +) + + +class BottleneckPatchEmbed(nn.Module): + # 3072 -> 1024 -> 4096 (raw 32x32 RGB patch -> bottleneck -> LLM hidden). + def __init__(self, patch_size=32, in_chans=3, pca_dim=1024, embed_dim=4096, bias=True, device=None, dtype=None, ops=None): + super().__init__() + self.proj1 = ops.Linear(patch_size * patch_size * in_chans, pca_dim, bias=False, device=device, dtype=dtype) + self.proj2 = ops.Linear(pca_dim, embed_dim, bias=bias, device=device, dtype=dtype) + + def forward(self, x): + return self.proj2(self.proj1(x)) + + +class FinalLayer(nn.Module): + # 4096 -> 3072 (LLM hidden -> flat pixel patch). + def __init__(self, hidden_size, patch_size=32, out_channels=3, device=None, dtype=None, ops=None): + super().__init__() + self.linear = ops.Linear(hidden_size, patch_size * patch_size * out_channels, bias=True, device=device, dtype=dtype) + + def forward(self, x): + return self.linear(x) + + +class HiDreamO1Transformer(nn.Module): + """HiDream-O1 unified pixel-level transformer.""" + + def __init__(self, image_model=None, dtype=None, device=None, operations=None, + text_config_overrides=None, vision_config_overrides=None, **kwargs): + super().__init__() + self.dtype = dtype + + text_cfg = HiDreamO1TextConfig(**(text_config_overrides or {})) + vision_cfg = dict(QWEN3VL_VISION_DEFAULTS) + if vision_config_overrides: + vision_cfg.update(vision_config_overrides) + vision_cfg["out_hidden_size"] = text_cfg.hidden_size + + self.text_config = text_cfg + self.vision_config = vision_cfg + self.hidden_size = text_cfg.hidden_size + self.patch_size = PATCH_SIZE + self.in_channels = 3 + self.tms_token_id = TMS_TOKEN_ID + + self.visual = Qwen35VisionModel(vision_cfg, device=device, dtype=dtype, ops=operations) + self.language_model = Llama2_(text_cfg, device=device, dtype=dtype, ops=operations) + self.t_embedder1 = TimestepEmbedder( + text_cfg.hidden_size, device=device, dtype=dtype, operations=operations, + ) + self.x_embedder = BottleneckPatchEmbed( + patch_size=self.patch_size, in_chans=self.in_channels, + pca_dim=text_cfg.hidden_size // 4, embed_dim=text_cfg.hidden_size, + bias=True, device=device, dtype=dtype, ops=operations, + ) + self.final_layer2 = FinalLayer( + text_cfg.hidden_size, patch_size=self.patch_size, + out_channels=self.in_channels, device=device, dtype=dtype, ops=operations, + ) + + self._visual_cache = None + self._kv_cache_entries = [] + + def clear_kv_cache(self): + self._kv_cache_entries = [] + self._visual_cache = None + + def forward(self, x, timesteps, context=None, transformer_options={}, **kwargs): + return comfy.patcher_extension.WrapperExecutor.new_class_executor( + self._forward, + self, + comfy.patcher_extension.get_all_wrappers(comfy.patcher_extension.WrappersMP.DIFFUSION_MODEL, transformer_options) + ).execute(x, timesteps, context, transformer_options, **kwargs) + + def _forward(self, x, timesteps, context=None, transformer_options={}, input_ids=None, attention_mask=None, position_ids=None, + vinput_mask=None, ar_len=None, ref_pixel_values=None, ref_image_grid_thw=None, ref_patches=None, **kwargs): + """Returns flow-match velocity (x - x_pred) / sigma""" + + if input_ids is None or position_ids is None: + raise ValueError("HiDreamO1Transformer requires input_ids and position_ids in conditioning") + + B, _, H, W = x.shape + h_p, w_p = H // self.patch_size, W // self.patch_size + tgt_image_len = h_p * w_p + + z = einops.rearrange( + x, 'B C (H p1) (W p2) -> B (H W) (C p1 p2)', + p1=self.patch_size, p2=self.patch_size, + ) + vinputs = torch.cat([z, ref_patches.to(z.dtype)], dim=1) if ref_patches is not None else z + + inputs_embeds = self.language_model.embed_tokens(input_ids).to(x.dtype) + + if ref_pixel_values is not None and ref_image_grid_thw is not None: + # ViT output is constant across sampling steps within a generation + # identity-key by the input tensor so refs don't recompute every step. + cached = self._visual_cache + if cached is not None and cached[0] is ref_pixel_values: + image_embeds = cached[1] + else: + ref_pv = ref_pixel_values.to(inputs_embeds.device) + ref_grid = ref_image_grid_thw.to(inputs_embeds.device).long() + # extra_conds wraps with a leading batch dim; refs are model-level so [0] always recovers them. + if ref_pv.dim() == 3: + ref_pv = ref_pv[0] + if ref_grid.dim() == 3: + ref_grid = ref_grid[0] + image_embeds = self.visual(ref_pv, ref_grid).to(inputs_embeds.dtype) + self._visual_cache = (ref_pixel_values, image_embeds) + # image_pad positions identical across batch (input_ids shared cond/uncond). + image_idx = (input_ids[0] == IMAGE_TOKEN_ID).nonzero(as_tuple=True)[0] + if image_idx.shape[0] != image_embeds.shape[0]: + raise ValueError( + f"Image-token count {image_idx.shape[0]} != ViT output count " + f"{image_embeds.shape[0]}; check tokenizer/processor alignment." + ) + inputs_embeds[:, image_idx] = image_embeds.unsqueeze(0).expand(B, -1, -1) + + sigma = timesteps.float() / 1000.0 + t_pixeldit = 1.0 - sigma + t_emb = self.t_embedder1(t_pixeldit * 1000, inputs_embeds.dtype) + tms_mask_3d = (input_ids == self.tms_token_id).unsqueeze(-1).expand_as(inputs_embeds) + inputs_embeds = torch.where(tms_mask_3d, t_emb.unsqueeze(1).expand_as(inputs_embeds), inputs_embeds) + + vinputs_embedded = self.x_embedder(vinputs.to(inputs_embeds.dtype)) + inputs_embeds = torch.cat([inputs_embeds, vinputs_embedded], dim=1) + + # extra_conds stores position_ids as (1, 3, T); process_cond repeats dim 0 to B. Take row 0. + freqs_cis = self.language_model.compute_freqs_cis(position_ids[0].to(x.device), x.device) + freqs_cis = tuple(t.to(x.dtype) for t in freqs_cis) + + two_pass_attn = make_two_pass_attention(ar_len, transformer_options=transformer_options) + patches_replace = transformer_options.get("patches_replace", {}) + blocks_replace = patches_replace.get("dit", {}) + transformer_options["total_blocks"] = len(self.language_model.layers) + transformer_options["block_type"] = "double" + + # Cache prefix K/V across steps. Key includes input_ids (prompt), ref_id + # (refs scatter into inputs_embeds), and position_ids (RoPE baked into cached K). + can_cache = not blocks_replace and ar_len > 0 + cache_len = ar_len if can_cache else 0 + ref_id = id(ref_pixel_values) if ref_pixel_values is not None else None + pos_ids_key = position_ids[..., :cache_len] if can_cache else position_ids + cache_entries = self._kv_cache_entries + # Drop stale entries from a previous device (model was unloaded and reloaded). + if cache_entries and cache_entries[0]["input_ids"].device != input_ids.device: + cache_entries = [] + self._kv_cache_entries = [] + kv_cache = None + if can_cache: + for entry in cache_entries: + ck = entry["input_ids"] + ep = entry["position_ids"] + if (entry["cache_len"] == cache_len + and ck.shape == input_ids.shape and torch.equal(ck, input_ids) + and entry["ref_id"] == ref_id + and ep.shape == pos_ids_key.shape and torch.equal(ep, pos_ids_key)): + kv_cache = entry + break + + if kv_cache is not None: + # Hot path: project Q/K/V only for fresh positions; past_key_value prepends cached AR K/V. + hidden_states = inputs_embeds[:, cache_len:] + sliced_freqs = tuple(t[..., cache_len:, :] for t in freqs_cis) + for i, layer in enumerate(self.language_model.layers): + transformer_options["block_index"] = i + K_i, V_i = kv_cache["kv"][i] + hidden_states, _ = layer( + x=hidden_states, attention_mask=None, freqs_cis=sliced_freqs, optimized_attention=two_pass_attn, + past_key_value=(K_i, V_i, cache_len), + ) + else: + # Cold path: run full sequence; if cacheable, snapshot K/V at AR positions. + snapshots = [] if can_cache else None + past_kv_cold = () if can_cache else None + hidden_states = inputs_embeds + for i, layer in enumerate(self.language_model.layers): + transformer_options["block_index"] = i + if ("double_block", i) in blocks_replace: + def block_wrap(args, _layer=layer): + out = {} + out["x"], _ = _layer( + x=args["x"], attention_mask=args.get("attention_mask"), + freqs_cis=args["freqs_cis"], optimized_attention=args["optimized_attention"], + past_key_value=None, + ) + return out + out = blocks_replace[("double_block", i)]( + {"x": hidden_states, "attention_mask": None, + "freqs_cis": freqs_cis, "optimized_attention": two_pass_attn, + "transformer_options": transformer_options}, + {"original_block": block_wrap}, + ) + hidden_states = out["x"] + else: + hidden_states, present_kv = layer( + x=hidden_states, attention_mask=None, + freqs_cis=freqs_cis, optimized_attention=two_pass_attn, + past_key_value=past_kv_cold, + ) + if snapshots is not None: + K, V, _ = present_kv + snapshots.append((K[:, :, :cache_len].contiguous(), + V[:, :, :cache_len].contiguous())) + if snapshots is not None: + # Cap at 2 entries (cond + uncond). Multi-cond workflows LRU-evict. + new_entry = { + "input_ids": input_ids.clone(), + "cache_len": cache_len, + "kv": snapshots, + "ref_id": ref_id, + "position_ids": pos_ids_key.clone(), + } + self._kv_cache_entries = (cache_entries + [new_entry])[-2:] + + if self.language_model.norm is not None: + hidden_states = self.language_model.norm(hidden_states) + + # Slice target-image positions before the final projection so the Linear only runs on tgt_image_len tokens. + # In the hot path hidden_states starts at original position cache_len, so masks/indices shift by cache_len. + sliced_offset = cache_len if kv_cache is not None else 0 + if vinput_mask is not None: + vmask = vinput_mask.to(x.device).bool() + if sliced_offset > 0: + vmask = vmask[:, sliced_offset:] + target_hidden = hidden_states[vmask].view(B, -1, hidden_states.shape[-1])[:, :tgt_image_len] + else: + txt_seq_len = input_ids.shape[1] + start = txt_seq_len - sliced_offset + target_hidden = hidden_states[:, start:start + tgt_image_len] + x_pred_tgt = self.final_layer2(target_hidden) + + # fp32 final subtraction, bf16 here noticeably degrades samples. + x_pred_img = einops.rearrange( + x_pred_tgt, 'B (H W) (C p1 p2) -> B C (H p1) (W p2)', + H=h_p, W=w_p, p1=self.patch_size, p2=self.patch_size, + ) + return (x.float() - x_pred_img.float()) / sigma.view(B, 1, 1, 1).clamp_min(1e-3) diff --git a/comfy/ldm/hidream_o1/utils.py b/comfy/ldm/hidream_o1/utils.py new file mode 100644 index 000000000..5a1249c72 --- /dev/null +++ b/comfy/ldm/hidream_o1/utils.py @@ -0,0 +1,173 @@ +"""HiDream-O1 input-prep helpers: image/resolution math and unified-sequence +RoPE position-id assembly. The fix_point offset in get_rope_index_fix_point +lets the target image and patchified ref images share spatial RoPE positions +despite living at different sequence indices — same 2D image plane. +""" + +import math +from typing import Optional + +import torch + + +PATCH_SIZE = 32 +CONDITION_IMAGE_SIZE = 384 # ViT-side base size for ref images + + +def resize_tensor(img_t, image_size, patch_size=16): + """img_t: (1, 3, H, W) float [0, 1]. Fit to image_size**2 area, patch-aligned, center-cropped.""" + + while min(img_t.shape[-2], img_t.shape[-1]) >= 2 * image_size: # Pre-halves with 2x2 box averaging while the image is still very large + img_t = torch.nn.functional.avg_pool2d(img_t, kernel_size=2, stride=2) + + _, _, height, width = img_t.shape + m = patch_size + s_max = image_size * image_size + scale = math.sqrt(s_max / (width * height)) + + candidates = [ + (round(width * scale) // m * m, round(height * scale) // m * m), + (round(width * scale) // m * m, math.floor(height * scale) // m * m), + (math.floor(width * scale) // m * m, round(height * scale) // m * m), + (math.floor(width * scale) // m * m, math.floor(height * scale) // m * m), + ] + candidates = sorted(candidates, key=lambda x: x[0] * x[1], reverse=True) + new_size = candidates[-1] + for c in candidates: + if c[0] * c[1] <= s_max: + new_size = c + break + + new_w, new_h = new_size + s1 = width / new_w + s2 = height / new_h + if s1 < s2: + resize_w, resize_h = new_w, round(height / s1) + else: + resize_w, resize_h = round(width / s2), new_h + img_t = torch.nn.functional.interpolate(img_t, size=(resize_h, resize_w), mode="bicubic") + top = (resize_h - new_h) // 2 + left = (resize_w - new_w) // 2 + return img_t[..., top:top + new_h, left:left + new_w] + + +def calculate_dimensions(max_size, ratio): + """(W, H) for an aspect ratio fitting in max_size**2 area, 32-aligned.""" + width = math.sqrt(max_size * max_size * ratio) + height = width / ratio + width = int(width / 32) * 32 + height = int(height / 32) * 32 + return width, height + + +def ref_max_size(target_max_dim, k): + """K-dependent ref-image max dim before patchifying.""" + if k == 1: + return target_max_dim + if k == 2: + return target_max_dim * 48 // 64 + if k <= 4: + return target_max_dim // 2 + if k <= 8: + return target_max_dim * 24 // 64 + return target_max_dim // 4 + + +def cond_image_size(k): + """K-dependent ViT-side image size.""" + if k <= 4: + return CONDITION_IMAGE_SIZE + if k <= 8: + return CONDITION_IMAGE_SIZE * 48 // 64 + return CONDITION_IMAGE_SIZE // 2 + + +def get_rope_index_fix_point( + spatial_merge_size: int, + image_token_id: int, + vision_start_token_id: int, + input_ids: Optional[torch.LongTensor] = None, + image_grid_thw: Optional[torch.LongTensor] = None, + attention_mask: Optional[torch.Tensor] = None, + skip_vision_start_token=None, + fix_point: int = 4096, +): + mrope_position_deltas = [] + if input_ids is not None and image_grid_thw is not None: + total_input_ids = input_ids + if attention_mask is None: + attention_mask = torch.ones_like(total_input_ids) + position_ids = torch.ones( + 3, input_ids.shape[0], input_ids.shape[1], + dtype=input_ids.dtype, device=input_ids.device, + ) + attention_mask = attention_mask.to(total_input_ids.device) + for i, input_ids_b in enumerate(total_input_ids): + fp = fix_point + image_index = 0 + input_ids_b = input_ids_b[attention_mask[i] == 1] + vision_start_indices = torch.argwhere(input_ids_b == vision_start_token_id).squeeze(1) + vision_tokens = input_ids_b[vision_start_indices + 1] + image_nums = (vision_tokens == image_token_id).sum() + input_tokens = input_ids_b.tolist() + llm_pos_ids_list = [] + st = 0 + remain_images = image_nums + for _ in range(image_nums): + if image_token_id in input_tokens and remain_images > 0: + ed = input_tokens.index(image_token_id, st) + else: + ed = len(input_tokens) + 1 + t = image_grid_thw[image_index][0] + h = image_grid_thw[image_index][1] + w = image_grid_thw[image_index][2] + image_index += 1 + remain_images -= 1 + llm_grid_t = t.item() + llm_grid_h = h.item() // spatial_merge_size + llm_grid_w = w.item() // spatial_merge_size + text_len = ed - st + text_len -= skip_vision_start_token[image_index - 1] + text_len = max(0, text_len) + st_idx = llm_pos_ids_list[-1].max() + 1 if len(llm_pos_ids_list) > 0 else 0 + llm_pos_ids_list.append(torch.arange(text_len).view(1, -1).expand(3, -1) + st_idx) + + t_index = torch.arange(llm_grid_t).view(-1, 1).expand(-1, llm_grid_h * llm_grid_w).flatten() + h_index = torch.arange(llm_grid_h).view(1, -1, 1).expand(llm_grid_t, -1, llm_grid_w).flatten() + w_index = torch.arange(llm_grid_w).view(1, 1, -1).expand(llm_grid_t, llm_grid_h, -1).flatten() + + if skip_vision_start_token[image_index - 1]: + if fp > 0: + fp = fp - st_idx + llm_pos_ids_list.append(torch.stack([t_index, h_index, w_index]) + fp + st_idx) + fp = 0 + else: + llm_pos_ids_list.append(torch.stack([t_index, h_index, w_index]) + text_len + st_idx) + st = ed + llm_grid_t * llm_grid_h * llm_grid_w + + if st < len(input_tokens): + st_idx = llm_pos_ids_list[-1].max() + 1 if len(llm_pos_ids_list) > 0 else 0 + text_len = len(input_tokens) - st + llm_pos_ids_list.append(torch.arange(text_len).view(1, -1).expand(3, -1) + st_idx) + + llm_positions = torch.cat(llm_pos_ids_list, dim=1).reshape(3, -1) + position_ids[..., i, attention_mask[i] == 1] = llm_positions.to(position_ids.device) + mrope_position_deltas.append(llm_positions.max() + 1 - len(total_input_ids[i])) + mrope_position_deltas = torch.tensor(mrope_position_deltas, device=input_ids.device).unsqueeze(1) + return position_ids, mrope_position_deltas + + if attention_mask is not None: + position_ids = attention_mask.long().cumsum(-1) - 1 + position_ids.masked_fill_(attention_mask == 0, 1) + position_ids = position_ids.unsqueeze(0).expand(3, -1, -1).to(attention_mask.device) + max_position_ids = position_ids.max(0, keepdim=False)[0].max(-1, keepdim=True)[0] + mrope_position_deltas = max_position_ids + 1 - attention_mask.shape[-1] + else: + position_ids = ( + torch.arange(input_ids.shape[1], device=input_ids.device) + .view(1, 1, -1).expand(3, input_ids.shape[0], -1) + ) + mrope_position_deltas = torch.zeros( + [input_ids.shape[0], 1], device=input_ids.device, dtype=input_ids.dtype, + ) + return position_ids, mrope_position_deltas diff --git a/comfy/lora.py b/comfy/lora.py index db8f16bcb..f11e26ec9 100644 --- a/comfy/lora.py +++ b/comfy/lora.py @@ -97,12 +97,14 @@ def load_lora(lora, to_load, log_missing=True): def model_lora_keys_clip(model, key_map={}): sdk = model.state_dict().keys() + prefix_set = set() for k in sdk: if k.endswith(".weight"): key_map["text_encoders.{}".format(k[:-len(".weight")])] = k #generic lora format without any weird key names tp = k.find(".transformer.") #also map without wrapper prefix for composite text encoder models if tp > 0 and not k.startswith("clip_"): key_map["text_encoders.{}".format(k[tp + 1:-len(".weight")])] = k + prefix_set.add(k.split('.')[0]) text_model_lora_key = "lora_te_text_model_encoder_layers_{}_{}" clip_l_present = False @@ -163,6 +165,13 @@ def model_lora_keys_clip(model, key_map={}): lora_key = "lora_te1_{}".format(l_key.replace(".", "_")) key_map[lora_key] = k + if len(prefix_set) == 1: + full_prefix = "{}.transformer.model.".format(next(iter(prefix_set))) # kohya anima and maybe other single TE models that use a single llama arch based te + for k in sdk: + if k.endswith(".weight"): + if k.startswith(full_prefix): + l_key = k[len(full_prefix):-len(".weight")] + key_map["lora_te_{}".format(l_key.replace(".", "_"))] = k k = "clip_g.transformer.text_projection.weight" if k in sdk: diff --git a/comfy/model_base.py b/comfy/model_base.py index dbed239e5..0736321b3 100644 --- a/comfy/model_base.py +++ b/comfy/model_base.py @@ -58,6 +58,8 @@ import comfy.ldm.cogvideo.model import comfy.ldm.rt_detr.rtdetr_v4 import comfy.ldm.ernie.model import comfy.ldm.sam3.detector +import comfy.ldm.hidream_o1.model +from comfy.ldm.hidream_o1.conditioning import build_extra_conds import comfy.model_management import comfy.patcher_extension @@ -1674,6 +1676,32 @@ class HiDream(BaseModel): out['image_cond'] = comfy.conds.CONDNoiseShape(self.process_latent_in(image_cond)) return out +class HiDreamO1(BaseModel): + """HiDream-O1-Image: pixel-space DiT (no VAE). Refs from HiDreamO1ReferenceImages and tokens from the stub TE flow through + extra_conds; the heavy preprocessing lives in comfy.ldm.hidream_o1.conditioning.""" + PATCH_SIZE = 32 + + def __init__(self, model_config, model_type=ModelType.FLOW, device=None): + super().__init__(model_config, model_type, device=device, unet_model=comfy.ldm.hidream_o1.model.HiDreamO1Transformer) + + def extra_conds(self, **kwargs): + out = super().extra_conds(**kwargs) + text_input_ids = kwargs.get("text_input_ids", None) + noise = kwargs.get("noise", None) + if text_input_ids is None or noise is None: + return out + + conds = build_extra_conds( + text_input_ids, noise, + ref_images=kwargs.get("reference_latents", None), + target_patch_size=self.PATCH_SIZE, + ) + for k, v in conds.items(): + # ar_len is a Python int (precomputed to avoid a GPU sync in forward). + cls = comfy.conds.CONDConstant if k == "ar_len" else comfy.conds.CONDRegular + out[k] = cls(v) + return out + class Chroma(Flux): def __init__(self, model_config, model_type=ModelType.FLUX, device=None, unet_model=comfy.ldm.chroma.model.Chroma): super().__init__(model_config, model_type, device=device, unet_model=unet_model) diff --git a/comfy/model_detection.py b/comfy/model_detection.py index 8ae456481..bc0b933bc 100644 --- a/comfy/model_detection.py +++ b/comfy/model_detection.py @@ -620,6 +620,9 @@ def detect_unet_config(state_dict, key_prefix, metadata=None): dit_config["guidance_cond_proj_dim"] = None#f"{key_prefix}t_embedder.cond_proj.weight" in state_dict_keys return dit_config + if '{}t_embedder1.mlp.0.weight'.format(key_prefix) in state_dict_keys and '{}x_embedder.proj1.weight'.format(key_prefix) in state_dict_keys: # HiDream-O1 + return {"image_model": "hidream_o1"} + if '{}caption_projection.0.linear.weight'.format(key_prefix) in state_dict_keys: # HiDream dit_config = {} dit_config["image_model"] = "hidream" diff --git a/comfy/model_patcher.py b/comfy/model_patcher.py index 33bdedfb1..2ea14bc2c 100644 --- a/comfy/model_patcher.py +++ b/comfy/model_patcher.py @@ -242,6 +242,37 @@ class LazyCastingParam(torch.nn.Parameter): return self.model.patch_weight_to_device(self.key, device_to=self.model.load_device, return_weight=True).to("cpu") +class LazyCastingQuantizedParam: + def __init__(self, model, key): + self.model = model + self.key = key + self.cpu_state_dict = None + + def state_dict_tensor(self, state_dict_key): + if self.cpu_state_dict is None: + weight = self.model.patch_weight_to_device(self.key, device_to=self.model.load_device, return_weight=True) + self.cpu_state_dict = {k: v.to("cpu") for k, v in weight.state_dict(self.key).items()} + return self.cpu_state_dict[state_dict_key] + + +class LazyCastingParamPiece(torch.nn.Parameter): + def __new__(cls, caster, state_dict_key, tensor): + return super().__new__(cls, tensor) + + def __init__(self, caster, state_dict_key, tensor): + self.caster = caster + self.state_dict_key = state_dict_key + + @property + def device(self): + return CustomTorchDevice + + def to(self, *args, **kwargs): + caster = self.caster + del self.caster + return caster.state_dict_tensor(self.state_dict_key) + + class ModelPatcher: def __init__(self, model, load_device, offload_device, size=0, weight_inplace_update=False): self.size = size @@ -1463,20 +1494,37 @@ class ModelPatcher: self.clear_cached_hook_weights() def state_dict_for_saving(self, clip_state_dict=None, vae_state_dict=None, clip_vision_state_dict=None): - unet_state_dict = self.model.diffusion_model.state_dict() - for k, v in unet_state_dict.items(): + original_state_dict = self.model.diffusion_model.state_dict() + unet_state_dict = {} + keys = list(original_state_dict) + while len(keys) > 0: + k = keys.pop(0) + v = original_state_dict[k] op_keys = k.rsplit('.', 1) if (len(op_keys) < 2) or op_keys[1] not in ["weight", "bias"]: + unet_state_dict[k] = v continue try: op = comfy.utils.get_attr(self.model.diffusion_model, op_keys[0]) except: + unet_state_dict[k] = v continue if not op or not hasattr(op, "comfy_cast_weights") or \ (hasattr(op, "comfy_patched_weights") and op.comfy_patched_weights == True): + unet_state_dict[k] = v continue key = "diffusion_model." + k - unet_state_dict[k] = LazyCastingParam(self, key, comfy.utils.get_attr(self.model, key)) + weight = comfy.utils.get_attr(self.model, key) + if isinstance(weight, QuantizedTensor) and k in original_state_dict: + qt_state_dict = weight.state_dict(k) + caster = LazyCastingQuantizedParam(self, key) + for group_key in (x for x in qt_state_dict if x in original_state_dict): + if group_key in keys: + keys.remove(group_key) + unet_state_dict.pop(group_key, "") + unet_state_dict[group_key] = LazyCastingParamPiece(caster, "diffusion_model." + group_key, original_state_dict[group_key]) + continue + unet_state_dict[k] = LazyCastingParam(self, key, weight) return self.model.state_dict_for_saving(unet_state_dict, clip_state_dict=clip_state_dict, vae_state_dict=vae_state_dict, clip_vision_state_dict=clip_vision_state_dict) def __del__(self): diff --git a/comfy/model_sampling.py b/comfy/model_sampling.py index cf2b5db5f..5af336e76 100644 --- a/comfy/model_sampling.py +++ b/comfy/model_sampling.py @@ -93,7 +93,8 @@ class CONST: def noise_scaling(self, sigma, noise, latent_image, max_denoise=False): sigma = reshape_sigma(sigma, noise.ndim) - return sigma * noise + (1.0 - sigma) * latent_image + s = getattr(self, "noise_scale", 1.0) + return sigma * (s * noise) + (1.0 - sigma) * latent_image def inverse_noise_scaling(self, sigma, latent): sigma = reshape_sigma(sigma, latent.ndim) @@ -288,7 +289,11 @@ class ModelSamplingDiscreteFlow(torch.nn.Module): else: sampling_settings = {} - self.set_parameters(shift=sampling_settings.get("shift", 1.0), multiplier=sampling_settings.get("multiplier", 1000)) + self.set_noise_scale(sampling_settings.get("noise_scale", 1.0)) + self.set_parameters( + shift=sampling_settings.get("shift", 1.0), + multiplier=sampling_settings.get("multiplier", 1000), + ) def set_parameters(self, shift=1.0, timesteps=1000, multiplier=1000): self.shift = shift @@ -296,6 +301,9 @@ class ModelSamplingDiscreteFlow(torch.nn.Module): ts = self.sigma((torch.arange(1, timesteps + 1, 1) / timesteps) * multiplier) self.register_buffer('sigmas', ts) + def set_noise_scale(self, noise_scale): + self.noise_scale = float(noise_scale) + @property def sigma_min(self): return self.sigmas[0] diff --git a/comfy/ops.py b/comfy/ops.py index 77ad1d527..117cdd327 100644 --- a/comfy/ops.py +++ b/comfy/ops.py @@ -1285,7 +1285,8 @@ def mixed_precision_ops(quant_config={}, compute_dtype=torch.bfloat16, full_prec if quant_format in ["float8_e4m3fn", "float8_e5m2"] and weight_key in state_dict: self.quant_format = quant_format qconfig = QUANT_ALGOS[quant_format] - layout_cls = get_layout_class(qconfig["comfy_tensor_layout"]) + self.layout_type = qconfig["comfy_tensor_layout"] + layout_cls = get_layout_class(self.layout_type) weight = state_dict.pop(weight_key) manually_loaded_keys = [weight_key] diff --git a/comfy/sd.py b/comfy/sd.py index 749bdd710..ab2718892 100644 --- a/comfy/sd.py +++ b/comfy/sd.py @@ -239,7 +239,8 @@ class CLIP: model_management.archive_model_dtypes(self.cond_stage_model) self.tokenizer = tokenizer(embedding_directory=embedding_directory, tokenizer_data=tokenizer_data) - ModelPatcher = comfy.model_patcher.ModelPatcher if disable_dynamic else comfy.model_patcher.CoreModelPatcher + te_disable_dynamic = disable_dynamic or getattr(self.cond_stage_model, "disable_offload", False) + ModelPatcher = comfy.model_patcher.ModelPatcher if te_disable_dynamic else comfy.model_patcher.CoreModelPatcher self.patcher = ModelPatcher(self.cond_stage_model, load_device=load_device, offload_device=offload_device) #Match torch.float32 hardcode upcast in TE implemention self.patcher.set_model_compute_dtype(torch.float32) @@ -776,6 +777,7 @@ class VAE: self.latent_channels = 3 self.latent_dim = 2 self.output_channels = 3 + self.disable_offload = True elif "vocoder.activation_post.downsample.lowpass.filter" in sd: #MMAudio VAE sample_rate = 16000 if sample_rate == 16000: diff --git a/comfy/supported_models.py b/comfy/supported_models.py index 40417f922..8d2e02f68 100644 --- a/comfy/supported_models.py +++ b/comfy/supported_models.py @@ -28,6 +28,7 @@ import comfy.text_encoders.ace15 import comfy.text_encoders.longcat_image import comfy.text_encoders.ernie import comfy.text_encoders.cogvideo +import comfy.text_encoders.hidream_o1 from . import supported_models_base from . import latent_formats @@ -1431,6 +1432,50 @@ class HiDream(supported_models_base.BASE): def clip_target(self, state_dict={}): return None # TODO +class HiDreamO1(supported_models_base.BASE): + unet_config = { + "image_model": "hidream_o1", + } + + sampling_settings = { + "shift": 3.0, + "noise_scale": 8.0, + } + + latent_format = latent_formats.HiDreamO1Pixel + memory_usage_factor = 0.6 + # fp16 not supported: LM MLP down_proj activations fp16 overflow, causing NaNs + supported_inference_dtypes = [torch.bfloat16, torch.float32] + + vae_key_prefix = ["vae."] + text_encoder_key_prefix = ["text_encoders."] + + optimizations = {"fp8": False} + + def get_model(self, state_dict, prefix="", device=None): + return model_base.HiDreamO1(self, device=device) + + def process_unet_state_dict(self, state_dict): + # Drop unused Qwen3-VL deepstack merger weights; upstream discards them at inference. + for key in list(state_dict.keys()): + if "visual.deepstack_merger_list" in key: + del state_dict[key] + return state_dict + + def process_vae_state_dict(self, state_dict): + # Pixel-space model: inject sentinel so VAE construction picks PixelspaceConversionVAE. + return {"pixel_space_vae": torch.tensor(1.0)} + + def process_clip_state_dict(self, state_dict): + # Tokenizer-only TE: inject sentinel so load_state_dict_guess_config triggers CLIP init. + return {"_hidream_o1_te_sentinel": torch.zeros(1)} + + def clip_target(self, state_dict={}): + return supported_models_base.ClipTarget( + comfy.text_encoders.hidream_o1.HiDreamO1Tokenizer, + comfy.text_encoders.hidream_o1.HiDreamO1TE, + ) + class Chroma(supported_models_base.BASE): unet_config = { "image_model": "chroma", @@ -2018,6 +2063,7 @@ models = [ Hunyuan3Dv2, Hunyuan3Dv2_1, HiDream, + HiDreamO1, Chroma, ChromaRadiance, ACEStep, diff --git a/comfy/text_encoders/hidream_o1.py b/comfy/text_encoders/hidream_o1.py new file mode 100644 index 000000000..5d287b784 --- /dev/null +++ b/comfy/text_encoders/hidream_o1.py @@ -0,0 +1,119 @@ +"""HiDream-O1-Image tokenizer-only text encoder. + +The real Qwen3-VL backbone runs inside diffusion_model.* every step, so this +module just tokenizes the prompt into text_input_ids and emits them as +conditioning. Position ids / token_types / vinput_mask depend on target H/W +and are built later in model_base.HiDreamO1.extra_conds. +""" + +import os + +import torch +from transformers import Qwen2Tokenizer + +from comfy import sd1_clip + + +# Qwen3-VL special tokens +IM_START_ID = 151644 +IM_END_ID = 151645 +ASSISTANT_ID = 77091 +USER_ID = 872 +NEWLINE_ID = 198 +VISION_START_ID = 151652 +VISION_END_ID = 151653 +IMAGE_TOKEN_ID = 151655 +VIDEO_TOKEN_ID = 151656 +# HiDream-O1-specific tokens +BOI_TOKEN_ID = 151669 +BOR_TOKEN_ID = 151670 +EOR_TOKEN_ID = 151671 +BOT_TOKEN_ID = 151672 +TMS_TOKEN_ID = 151673 + + +class HiDreamO1QwenTokenizer(sd1_clip.SDTokenizer): + def __init__(self, embedding_directory=None, tokenizer_data={}): + tokenizer_path = os.path.join( + os.path.dirname(os.path.realpath(__file__)), "qwen25_tokenizer" + ) + super().__init__( + tokenizer_path, + pad_with_end=False, + embedding_size=4096, + embedding_key="hidream_o1", + tokenizer_class=Qwen2Tokenizer, + has_start_token=False, + has_end_token=False, + pad_to_max_length=False, + max_length=99999999, + min_length=1, + pad_token=151643, + tokenizer_data=tokenizer_data, + ) + + +class HiDreamO1Tokenizer(sd1_clip.SD1Tokenizer): + """Wraps prompt in the upstream chat template ending with boi/tms markers. + Image tokens get spliced in at sample time once target H/W is known. + """ + + def __init__(self, embedding_directory=None, tokenizer_data={}): + super().__init__( + embedding_directory=embedding_directory, + tokenizer_data=tokenizer_data, + name="hidream_o1", + tokenizer=HiDreamO1QwenTokenizer, + ) + + def tokenize_with_weights(self, text, return_word_ids=False, **kwargs): + text_tokens_dict = super().tokenize_with_weights( + text, return_word_ids=return_word_ids, disable_weights=True, **kwargs + ) + text_tuples = text_tokens_dict["hidream_o1"][0] + text_tuples = [t for t in text_tuples if int(t[0]) != 151643] # strip pad + + # <|im_start|>user\n{text}<|im_end|>\n<|im_start|>assistant\n<|boi|><|tms|> + def tok(tid): + return (tid, 1.0) if not return_word_ids else (tid, 1.0, 0) + + prefix = [tok(IM_START_ID), tok(USER_ID), tok(NEWLINE_ID)] + suffix = [ + tok(IM_END_ID), tok(NEWLINE_ID), + tok(IM_START_ID), tok(ASSISTANT_ID), tok(NEWLINE_ID), + tok(BOI_TOKEN_ID), tok(TMS_TOKEN_ID), + ] + full = prefix + list(text_tuples) + suffix + return {"hidream_o1": [full]} + + +class HiDreamO1TE(torch.nn.Module): + """Passthrough TE: emits int token ids; the Qwen3-VL backbone in diffusion_model does the actual encoding.""" + + def __init__(self, device="cpu", dtype=None, model_options={}): + super().__init__() + self.dtypes = {torch.float32} + self.disable_offload = True # skips dynamic VRAM management for this zero-parameter module + self.device = torch.device("cpu") if device is None else torch.device(device) + + def encode_token_weights(self, token_weight_pairs): + tok_pairs = token_weight_pairs["hidream_o1"][0] + ids = [int(t[0]) for t in tok_pairs] + input_ids = torch.tensor([ids], dtype=torch.long) + # Surrogate keeps the cross_attn slot non-empty for CONDITIONING + # plumbing; the model reads text_input_ids out of `extra` instead. + cross_attn = input_ids.unsqueeze(-1).to(torch.float32) + extra = {"text_input_ids": input_ids} + return cross_attn, None, extra + + def load_sd(self, sd): + return [] + + def get_sd(self): + return {} + + def reset_clip_options(self): + pass + + def set_clip_options(self, options): + pass diff --git a/comfy/text_encoders/llama.py b/comfy/text_encoders/llama.py index a34c41144..5087228ca 100644 --- a/comfy/text_encoders/llama.py +++ b/comfy/text_encoders/llama.py @@ -397,7 +397,7 @@ class RMSNorm(nn.Module): -def precompute_freqs_cis(head_dim, position_ids, theta, rope_scale=None, rope_dims=None, device=None): +def precompute_freqs_cis(head_dim, position_ids, theta, rope_scale=None, rope_dims=None, device=None, interleaved_mrope=False): if not isinstance(theta, list): theta = [theta] @@ -415,16 +415,27 @@ def precompute_freqs_cis(head_dim, position_ids, theta, rope_scale=None, rope_di inv_freq_expanded = inv_freq[None, :, None].float().expand(position_ids.shape[0], -1, 1) position_ids_expanded = position_ids[:, None, :].float() freqs = (inv_freq_expanded.float() @ position_ids_expanded.float()).transpose(1, 2) - emb = torch.cat((freqs, freqs), dim=-1) - cos = emb.cos() - sin = emb.sin() - if rope_dims is not None and position_ids.shape[0] > 1: - mrope_section = rope_dims * 2 - cos = torch.cat([m[i % 3] for i, m in enumerate(cos.split(mrope_section, dim=-1))], dim=-1).unsqueeze(0) - sin = torch.cat([m[i % 3] for i, m in enumerate(sin.split(mrope_section, dim=-1))], dim=-1).unsqueeze(0) + if rope_dims is not None and position_ids.shape[0] > 1 and interleaved_mrope: + # Qwen3-VL interleaved MRoPE: T-freqs by default, H/W replace every 3rd dim. + freqs_inter = freqs[0].clone() + for axis_idx, offset in ((1, 1), (2, 2)): + length = rope_dims[axis_idx] * 3 + idx = slice(offset, length, 3) + freqs_inter[..., idx] = freqs[axis_idx, ..., idx] + emb = torch.cat((freqs_inter, freqs_inter), dim=-1) + cos = emb.cos().unsqueeze(0) + sin = emb.sin().unsqueeze(0) else: - cos = cos.unsqueeze(1) - sin = sin.unsqueeze(1) + emb = torch.cat((freqs, freqs), dim=-1) + cos = emb.cos() + sin = emb.sin() + if rope_dims is not None and position_ids.shape[0] > 1: + mrope_section = rope_dims * 2 + cos = torch.cat([m[i % 3] for i, m in enumerate(cos.split(mrope_section, dim=-1))], dim=-1).unsqueeze(0) + sin = torch.cat([m[i % 3] for i, m in enumerate(sin.split(mrope_section, dim=-1))], dim=-1).unsqueeze(0) + else: + cos = cos.unsqueeze(1) + sin = sin.unsqueeze(1) sin_split = sin.shape[-1] // 2 out.append((cos, sin[..., : sin_split], -sin[..., sin_split :])) @@ -689,6 +700,7 @@ class Llama2_(nn.Module): self.config.rope_theta, self.config.rope_scale, self.config.rope_dims, + interleaved_mrope=getattr(self.config, "interleaved_mrope", False), device=device) def forward(self, x, attention_mask=None, embeds=None, num_tokens=None, intermediate_output=None, final_layer_norm_intermediate=True, dtype=None, position_ids=None, embeds_info=[], past_key_values=None, input_ids=None): diff --git a/comfy/text_encoders/qwen35.py b/comfy/text_encoders/qwen35.py index d8ed9cd32..b022009b1 100644 --- a/comfy/text_encoders/qwen35.py +++ b/comfy/text_encoders/qwen35.py @@ -451,9 +451,8 @@ class Qwen35VisionPatchEmbed(nn.Module): self.proj = ops.Conv3d(self.in_channels, self.embed_dim, kernel_size=kernel_size, stride=kernel_size, bias=True, device=device, dtype=dtype) def forward(self, x): - target_dtype = self.proj.weight.dtype x = x.view(-1, self.in_channels, self.temporal_patch_size, self.patch_size, self.patch_size) - return self.proj(x.to(target_dtype)).view(-1, self.embed_dim) + return self.proj(x).view(-1, self.embed_dim) class Qwen35VisionMLP(nn.Module): @@ -651,7 +650,7 @@ class Qwen35VisionModel(nn.Module): x = self.patch_embed(x) pos_embeds = self.fast_pos_embed_interpolate(grid_thw).to(x.device) x = x + pos_embeds - rotary_pos_emb = self.rot_pos_emb(grid_thw) + rotary_pos_emb = self.rot_pos_emb(grid_thw).to(x.device) seq_len = x.shape[0] x = x.reshape(seq_len, -1) rotary_pos_emb = rotary_pos_emb.reshape(seq_len, -1) diff --git a/comfy/utils.py b/comfy/utils.py index 91e1ba3d3..b75972027 100644 --- a/comfy/utils.py +++ b/comfy/utils.py @@ -1196,7 +1196,7 @@ def model_trange(*args, **kwargs): pbar.i1_time = time.time() pbar.set_postfix_str(" Model Initialization complete! ") elif pbar._i == 2: - #bring forward the effective start time based the the diff between first and second iteration + #bring forward the effective start time based the diff between first and second iteration #to attempt to remove load overhead from the final step rate estimate. pbar.start_t = pbar.i1_time - (time.time() - pbar.i1_time) pbar.set_postfix_str("") diff --git a/comfy_api_nodes/apis/bria.py b/comfy_api_nodes/apis/bria.py index 8c496b56c..e08a519a8 100644 --- a/comfy_api_nodes/apis/bria.py +++ b/comfy_api_nodes/apis/bria.py @@ -23,7 +23,7 @@ class BriaEditImageRequest(BaseModel): None, description="Mask image (black and white). Black areas will be preserved, white areas will be edited. " "If omitted, the edit applies to the entire image. " - "The input image and the the input mask must be of the same size.", + "The input image and the input mask must be of the same size.", ) negative_prompt: str | None = Field(None) guidance_scale: float = Field(...) diff --git a/comfy_api_nodes/apis/bytedance.py b/comfy_api_nodes/apis/bytedance.py index c05bd6893..03f4c445b 100644 --- a/comfy_api_nodes/apis/bytedance.py +++ b/comfy_api_nodes/apis/bytedance.py @@ -198,6 +198,62 @@ RECOMMENDED_PRESETS_SEEDREAM_4 = [ ("Custom", None, None), ] +_PRESETS_SEEDREAM_1K = [ + ("(1K) 1024x1024 (1:1)", 1024, 1024), + ("(1K) 864x1152 (3:4)", 864, 1152), + ("(1K) 1152x864 (4:3)", 1152, 864), + ("(1K) 1312x736 (16:9)", 1312, 736), + ("(1K) 736x1312 (9:16)", 736, 1312), + ("(1K) 832x1248 (2:3)", 832, 1248), + ("(1K) 1248x832 (3:2)", 1248, 832), + ("(1K) 1568x672 (21:9)", 1568, 672), +] + +_PRESETS_SEEDREAM_2K = [ + ("(2K) 2048x2048 (1:1)", 2048, 2048), + ("(2K) 1728x2304 (3:4)", 1728, 2304), + ("(2K) 2304x1728 (4:3)", 2304, 1728), + ("(2K) 2848x1600 (16:9)", 2848, 1600), + ("(2K) 1600x2848 (9:16)", 1600, 2848), + ("(2K) 1664x2496 (2:3)", 1664, 2496), + ("(2K) 2496x1664 (3:2)", 2496, 1664), + ("(2K) 3136x1344 (21:9)", 3136, 1344), +] + +_PRESETS_SEEDREAM_3K = [ + ("(3K) 3072x3072 (1:1)", 3072, 3072), + ("(3K) 2592x3456 (3:4)", 2592, 3456), + ("(3K) 3456x2592 (4:3)", 3456, 2592), + ("(3K) 4096x2304 (16:9)", 4096, 2304), + ("(3K) 2304x4096 (9:16)", 2304, 4096), + ("(3K) 2496x3744 (2:3)", 2496, 3744), + ("(3K) 3744x2496 (3:2)", 3744, 2496), + ("(3K) 4704x2016 (21:9)", 4704, 2016), +] + +_PRESETS_SEEDREAM_4K = [ + ("(4K) 4096x4096 (1:1)", 4096, 4096), + ("(4K) 3520x4704 (3:4)", 3520, 4704), + ("(4K) 4704x3520 (4:3)", 4704, 3520), + ("(4K) 5504x3040 (16:9)", 5504, 3040), + ("(4K) 3040x5504 (9:16)", 3040, 5504), + ("(4K) 3328x4992 (2:3)", 3328, 4992), + ("(4K) 4992x3328 (3:2)", 4992, 3328), + ("(4K) 6240x2656 (21:9)", 6240, 2656), +] + +_CUSTOM_PRESET = [("Custom", None, None)] + +RECOMMENDED_PRESETS_SEEDREAM_5_LITE = ( + _PRESETS_SEEDREAM_2K + _PRESETS_SEEDREAM_3K + _PRESETS_SEEDREAM_4K + _CUSTOM_PRESET +) +RECOMMENDED_PRESETS_SEEDREAM_4_5 = ( + _PRESETS_SEEDREAM_2K + _PRESETS_SEEDREAM_4K + _CUSTOM_PRESET +) +RECOMMENDED_PRESETS_SEEDREAM_4_0 = ( + _PRESETS_SEEDREAM_1K + _PRESETS_SEEDREAM_2K + _PRESETS_SEEDREAM_4K + _CUSTOM_PRESET +) + # Seedance 2.0 reference video pixel count limits per model and output resolution. SEEDANCE2_REF_VIDEO_PIXEL_LIMITS = { "dreamina-seedance-2-0-260128": { diff --git a/comfy_api_nodes/nodes_bfl.py b/comfy_api_nodes/nodes_bfl.py index 23590bf24..3f0ce29d8 100644 --- a/comfy_api_nodes/nodes_bfl.py +++ b/comfy_api_nodes/nodes_bfl.py @@ -596,6 +596,7 @@ class Flux2ProImageNode(IO.ComfyNode): depends_on=IO.PriceBadgeDepends(widgets=["width", "height"], inputs=["images"]), expr=cls.PRICE_BADGE_EXPR, ), + is_deprecated=True, ) @classmethod @@ -674,6 +675,175 @@ class Flux2MaxImageNode(Flux2ProImageNode): """ +_FLUX2_MODEL_ENDPOINTS = { + "Flux.2 [pro]": "/proxy/bfl/flux-2-pro/generate", + "Flux.2 [max]": "/proxy/bfl/flux-2-max/generate", +} + + +def _flux2_model_inputs(): + return [ + IO.Int.Input( + "width", + default=1024, + min=256, + max=2048, + step=32, + ), + IO.Int.Input( + "height", + default=768, + min=256, + max=2048, + step=32, + ), + IO.Autogrow.Input( + "images", + template=IO.Autogrow.TemplateNames( + IO.Image.Input("image"), + names=[f"image_{i}" for i in range(1, 9)], + min=0, + ), + tooltip="Optional reference image(s) for image-to-image generation. Up to 8 images.", + ), + ] + + +class Flux2ImageNode(IO.ComfyNode): + + @classmethod + def define_schema(cls) -> IO.Schema: + return IO.Schema( + node_id="Flux2ImageNode", + display_name="Flux.2 Image", + category="api node/image/BFL", + description="Generate images via Flux.2 [pro] or Flux.2 [max] from a prompt and optional reference images.", + inputs=[ + IO.String.Input( + "prompt", + multiline=True, + default="", + tooltip="Prompt for the image generation or edit", + ), + IO.DynamicCombo.Input( + "model", + options=[ + IO.DynamicCombo.Option("Flux.2 [pro]", _flux2_model_inputs()), + IO.DynamicCombo.Option("Flux.2 [max]", _flux2_model_inputs()), + ], + ), + IO.Int.Input( + "seed", + default=0, + min=0, + max=0xFFFFFFFFFFFFFFFF, + control_after_generate=True, + tooltip="The random seed used for creating the noise.", + ), + ], + outputs=[IO.Image.Output()], + hidden=[ + IO.Hidden.auth_token_comfy_org, + IO.Hidden.api_key_comfy_org, + IO.Hidden.unique_id, + ], + is_api_node=True, + price_badge=IO.PriceBadge( + depends_on=IO.PriceBadgeDepends( + widgets=["model", "model.width", "model.height"], + input_groups=["model.images"], + ), + expr=""" + ( + $isMax := widgets.model = "flux.2 [max]"; + $MP := 1024 * 1024; + $w := $lookup(widgets, "model.width"); + $h := $lookup(widgets, "model.height"); + $outMP := $max([1, $floor((($w * $h) + $MP - 1) / $MP)]); + $outputCost := $isMax + ? (0.07 + 0.03 * ($outMP - 1)) + : (0.03 + 0.015 * ($outMP - 1)); + $refMin := $isMax ? 0.03 : 0.015; + $refMax := $isMax ? 0.24 : 0.12; + $hasRefs := $lookup(inputGroups, "model.images") > 0; + $hasRefs + ? { + "type": "range_usd", + "min_usd": $outputCost + $refMin, + "max_usd": $outputCost + $refMax, + "format": { "approximate": true } + } + : {"type": "usd", "usd": $outputCost} + ) + """, + ), + ) + + @classmethod + async def execute( + cls, + prompt: str, + model: dict, + seed: int, + ) -> IO.NodeOutput: + model_choice = model["model"] + endpoint = _FLUX2_MODEL_ENDPOINTS[model_choice] + width = model["width"] + height = model["height"] + images_dict = model.get("images") or {} + + image_tensors: list[Input.Image] = [t for t in images_dict.values() if t is not None] + n_images = sum(get_number_of_images(t) for t in image_tensors) + if n_images > 8: + raise ValueError("The current maximum number of supported images is 8.") + + flat_tensors: list[torch.Tensor] = [] + for tensor in image_tensors: + if len(tensor.shape) == 4: + flat_tensors.extend(tensor[i] for i in range(tensor.shape[0])) + else: + flat_tensors.append(tensor) + + reference_images: dict[str, str] = {} + for idx, tensor in enumerate(flat_tensors): + key_name = f"input_image_{idx + 1}" if idx else "input_image" + reference_images[key_name] = tensor_to_base64_string(tensor, total_pixels=2048 * 2048) + + initial_response = await sync_op( + cls, + ApiEndpoint(path=endpoint, method="POST"), + response_model=BFLFluxProGenerateResponse, + data=Flux2ProGenerateRequest( + prompt=prompt, + width=width, + height=height, + seed=seed, + **reference_images, + ), + ) + + def price_extractor(_r: BaseModel) -> float | None: + return None if initial_response.cost is None else initial_response.cost / 100 + + response = await poll_op( + cls, + ApiEndpoint(initial_response.polling_url), + response_model=BFLFluxStatusResponse, + status_extractor=lambda r: r.status, + progress_extractor=lambda r: r.progress, + price_extractor=price_extractor, + completed_statuses=[BFLStatus.ready], + failed_statuses=[ + BFLStatus.request_moderated, + BFLStatus.content_moderated, + BFLStatus.error, + BFLStatus.task_not_found, + ], + queued_statuses=[], + ) + return IO.NodeOutput(await download_url_to_image_tensor(response.result["sample"])) + + class BFLExtension(ComfyExtension): @override async def get_node_list(self) -> list[type[IO.ComfyNode]]: @@ -685,6 +855,7 @@ class BFLExtension(ComfyExtension): FluxProFillNode, Flux2ProImageNode, Flux2MaxImageNode, + Flux2ImageNode, ] diff --git a/comfy_api_nodes/nodes_bytedance.py b/comfy_api_nodes/nodes_bytedance.py index 5f74f4a14..d6b479336 100644 --- a/comfy_api_nodes/nodes_bytedance.py +++ b/comfy_api_nodes/nodes_bytedance.py @@ -10,6 +10,9 @@ from comfy_api.latest import IO, ComfyExtension, Input from comfy_api_nodes.apis.bytedance import ( RECOMMENDED_PRESETS, RECOMMENDED_PRESETS_SEEDREAM_4, + RECOMMENDED_PRESETS_SEEDREAM_4_0, + RECOMMENDED_PRESETS_SEEDREAM_4_5, + RECOMMENDED_PRESETS_SEEDREAM_5_LITE, SEEDANCE2_PRICE_PER_1K_TOKENS, SEEDANCE2_REF_VIDEO_PIXEL_LIMITS, VIDEO_TASKS_EXECUTION_TIME, @@ -68,6 +71,12 @@ SEEDREAM_MODELS = { "seedream-4-0-250828": "seedream-4-0-250828", } +SEEDREAM_PRESETS = { + "seedream-5-0-260128": RECOMMENDED_PRESETS_SEEDREAM_5_LITE, + "seedream-4-5-251128": RECOMMENDED_PRESETS_SEEDREAM_4_5, + "seedream-4-0-250828": RECOMMENDED_PRESETS_SEEDREAM_4_0, +} + # Long-running tasks endpoints(e.g., video) BYTEPLUS_TASK_ENDPOINT = "/proxy/byteplus/api/v3/contents/generations/tasks" BYTEPLUS_TASK_STATUS_ENDPOINT = "/proxy/byteplus/api/v3/contents/generations/tasks" # + /{task_id} @@ -562,6 +571,7 @@ class ByteDanceSeedreamNode(IO.ComfyNode): ) """, ), + is_deprecated=True, ) @classmethod @@ -651,6 +661,226 @@ class ByteDanceSeedreamNode(IO.ComfyNode): return IO.NodeOutput(torch.cat([await download_url_to_image_tensor(i) for i in urls])) +def _seedream_model_inputs(*, max_ref_images: int, presets: list): + return [ + IO.Combo.Input( + "size_preset", + options=[label for label, _, _ in presets], + tooltip="Pick a recommended size. Select Custom to use the width and height below.", + ), + IO.Int.Input( + "width", + default=2048, + min=1024, + max=6240, + step=2, + tooltip="Custom width for image. Value is working only if `size_preset` is set to `Custom`", + ), + IO.Int.Input( + "height", + default=2048, + min=1024, + max=4992, + step=2, + tooltip="Custom height for image. Value is working only if `size_preset` is set to `Custom`", + ), + IO.Int.Input( + "max_images", + default=1, + min=1, + max=max_ref_images, + step=1, + display_mode=IO.NumberDisplay.number, + tooltip="Maximum number of images to generate. With 1, exactly one image is produced. " + "With >1, the model generates between 1 and max_images related images " + "(e.g., story scenes, character variations). " + "Total images (input + generated) cannot exceed 15.", + ), + IO.Autogrow.Input( + "images", + template=IO.Autogrow.TemplateNames( + IO.Image.Input("image"), + names=[f"image_{i}" for i in range(1, max_ref_images + 1)], + min=0, + ), + tooltip=f"Optional reference image(s) for image-to-image or multi-reference generation. " + f"Up to {max_ref_images} images.", + ), + IO.Boolean.Input( + "fail_on_partial", + default=False, + tooltip="If enabled, abort execution if any requested images are missing or return an error.", + advanced=True, + ), + ] + + +class ByteDanceSeedreamNodeV2(IO.ComfyNode): + + @classmethod + def define_schema(cls): + return IO.Schema( + node_id="ByteDanceSeedreamNodeV2", + display_name="ByteDance Seedream 4.5 & 5.0", + category="api node/image/ByteDance", + description="Unified text-to-image generation and precise single-sentence editing at up to 4K resolution.", + inputs=[ + IO.String.Input( + "prompt", + multiline=True, + default="", + tooltip="Text prompt for creating or editing an image.", + ), + IO.DynamicCombo.Input( + "model", + options=[ + IO.DynamicCombo.Option( + "seedream 5.0 lite", + _seedream_model_inputs(max_ref_images=14, presets=RECOMMENDED_PRESETS_SEEDREAM_5_LITE), + ), + IO.DynamicCombo.Option( + "seedream-4-5-251128", + _seedream_model_inputs(max_ref_images=10, presets=RECOMMENDED_PRESETS_SEEDREAM_4_5), + ), + IO.DynamicCombo.Option( + "seedream-4-0-250828", + _seedream_model_inputs(max_ref_images=10, presets=RECOMMENDED_PRESETS_SEEDREAM_4_0), + ), + ], + ), + IO.Int.Input( + "seed", + default=0, + min=0, + max=2147483647, + step=1, + display_mode=IO.NumberDisplay.number, + control_after_generate=True, + tooltip="Seed to use for generation.", + ), + IO.Boolean.Input( + "watermark", + default=False, + tooltip='Whether to add an "AI generated" watermark to the image.', + advanced=True, + ), + ], + outputs=[ + IO.Image.Output(), + ], + hidden=[ + IO.Hidden.auth_token_comfy_org, + IO.Hidden.api_key_comfy_org, + IO.Hidden.unique_id, + ], + is_api_node=True, + price_badge=IO.PriceBadge( + depends_on=IO.PriceBadgeDepends(widgets=["model"]), + expr=""" + ( + $price := $contains(widgets.model, "5.0 lite") ? 0.035 : + $contains(widgets.model, "4-5") ? 0.04 : 0.03; + { + "type":"usd", + "usd": $price, + "format": { "suffix":" x images/Run", "approximate": true } + } + ) + """, + ), + ) + + @classmethod + async def execute( + cls, + prompt: str, + model: dict, + seed: int = 0, + watermark: bool = False, + ) -> IO.NodeOutput: + validate_string(prompt, strip_whitespace=True, min_length=1) + model_id = SEEDREAM_MODELS[model["model"]] + presets = SEEDREAM_PRESETS[model_id] + + size_preset = model.get("size_preset", presets[0][0]) + width = model.get("width", 2048) + height = model.get("height", 2048) + max_images = model.get("max_images", 1) + sequential_image_generation = "disabled" if max_images == 1 else "auto" + images_dict = model.get("images") or {} + fail_on_partial = model.get("fail_on_partial", False) + + w = h = None + for label, tw, th in presets: + if label == size_preset: + w, h = tw, th + break + if w is None or h is None: + w, h = width, height + + out_num_pixels = w * h + mp_provided = out_num_pixels / 1_000_000.0 + if ("seedream-4-5" in model_id or "seedream-5-0" in model_id) and out_num_pixels < 3686400: + raise ValueError( + f"Minimum image resolution for the selected model is 3.68MP, but {mp_provided:.2f}MP provided." + ) + if "seedream-4-0" in model_id and out_num_pixels < 921600: + raise ValueError( + f"Minimum image resolution that the selected model can generate is 0.92MP, " + f"but {mp_provided:.2f}MP provided." + ) + if out_num_pixels > 16_777_216: + raise ValueError( + f"Maximum image resolution for the selected model is 16.78MP, but {mp_provided:.2f}MP provided." + ) + + image_tensors: list[Input.Image] = [t for t in images_dict.values() if t is not None] + n_input_images = sum(get_number_of_images(t) for t in image_tensors) + max_num_of_images = 14 if model_id == "seedream-5-0-260128" else 10 + if n_input_images > max_num_of_images: + raise ValueError( + f"Maximum of {max_num_of_images} reference images are supported, but {n_input_images} received." + ) + if sequential_image_generation == "auto" and n_input_images + max_images > 15: + raise ValueError( + "The maximum number of generated images plus the number of reference images cannot exceed 15." + ) + + reference_images_urls: list[str] = [] + if image_tensors: + for tensor in image_tensors: + validate_image_aspect_ratio(tensor, (1, 3), (3, 1)) + reference_images_urls = await upload_images_to_comfyapi( + cls, + image_tensors, + max_images=n_input_images, + mime_type="image/png", + wait_label="Uploading reference images", + ) + + response = await sync_op( + cls, + ApiEndpoint(path=BYTEPLUS_IMAGE_ENDPOINT, method="POST"), + response_model=ImageTaskCreationResponse, + data=Seedream4TaskCreationRequest( + model=model_id, + prompt=prompt, + image=reference_images_urls, + size=f"{w}x{h}", + seed=seed, + sequential_image_generation=sequential_image_generation, + sequential_image_generation_options=Seedream4Options(max_images=max_images), + watermark=watermark, + ), + ) + if len(response.data) == 1: + return IO.NodeOutput(await download_url_to_image_tensor(get_image_url_from_response(response))) + urls = [str(d["url"]) for d in response.data if isinstance(d, dict) and "url" in d] + if fail_on_partial and len(urls) < len(response.data): + raise RuntimeError(f"Only {len(urls)} of {len(response.data)} images were generated before error.") + return IO.NodeOutput(torch.cat([await download_url_to_image_tensor(i) for i in urls])) + + class ByteDanceTextToVideoNode(IO.ComfyNode): @classmethod @@ -2105,6 +2335,7 @@ class ByteDanceExtension(ComfyExtension): return [ ByteDanceImageNode, ByteDanceSeedreamNode, + ByteDanceSeedreamNodeV2, ByteDanceTextToVideoNode, ByteDanceImageToVideoNode, ByteDanceFirstLastFrameNode, diff --git a/comfy_api_nodes/nodes_grok.py b/comfy_api_nodes/nodes_grok.py index dd5d7e249..a103f24ee 100644 --- a/comfy_api_nodes/nodes_grok.py +++ b/comfy_api_nodes/nodes_grok.py @@ -162,6 +162,61 @@ class GrokImageNode(IO.ComfyNode): ) +_GROK_IMAGE_EDIT_ASPECT_RATIO_OPTIONS = [ + "auto", + "1:1", + "2:3", + "3:2", + "3:4", + "4:3", + "9:16", + "16:9", + "9:19.5", + "19.5:9", + "9:20", + "20:9", + "1:2", + "2:1", +] + + +def _grok_image_edit_model_inputs(*, max_ref_images: int, with_aspect_ratio: bool): + inputs = [ + IO.Autogrow.Input( + "images", + template=IO.Autogrow.TemplateNames( + IO.Image.Input("image"), + names=[f"image_{i}" for i in range(1, max_ref_images + 1)], + min=1, + ), + tooltip=( + "Reference image to edit." + if max_ref_images == 1 + else f"Reference image(s) to edit. Up to {max_ref_images} images." + ), + ), + IO.Combo.Input("resolution", options=["1K", "2K"]), + IO.Int.Input( + "number_of_images", + default=1, + min=1, + max=10, + step=1, + tooltip="Number of edited images to generate", + display_mode=IO.NumberDisplay.number, + ), + ] + if with_aspect_ratio: + inputs.append( + IO.Combo.Input( + "aspect_ratio", + options=_GROK_IMAGE_EDIT_ASPECT_RATIO_OPTIONS, + tooltip="Only allowed when multiple images are connected.", + ) + ) + return inputs + + class GrokImageEditNode(IO.ComfyNode): @classmethod @@ -256,6 +311,7 @@ class GrokImageEditNode(IO.ComfyNode): ) """, ), + is_deprecated=True, ) @classmethod @@ -303,6 +359,143 @@ class GrokImageEditNode(IO.ComfyNode): ) +class GrokImageEditNodeV2(IO.ComfyNode): + + @classmethod + def define_schema(cls): + return IO.Schema( + node_id="GrokImageEditNodeV2", + display_name="Grok Image Edit", + category="api node/image/Grok", + description="Modify an existing image based on a text prompt", + inputs=[ + IO.String.Input( + "prompt", + multiline=True, + default="", + tooltip="The text prompt used to generate the image", + ), + IO.DynamicCombo.Input( + "model", + options=[ + IO.DynamicCombo.Option( + "grok-imagine-image-quality", + _grok_image_edit_model_inputs(max_ref_images=3, with_aspect_ratio=True), + ), + IO.DynamicCombo.Option( + "grok-imagine-image-pro", + _grok_image_edit_model_inputs(max_ref_images=1, with_aspect_ratio=False), + ), + IO.DynamicCombo.Option( + "grok-imagine-image", + _grok_image_edit_model_inputs(max_ref_images=3, with_aspect_ratio=True), + ), + ], + ), + IO.Int.Input( + "seed", + default=0, + min=0, + max=2147483647, + step=1, + display_mode=IO.NumberDisplay.number, + control_after_generate=True, + tooltip="Seed to determine if node should re-run; " + "actual results are nondeterministic regardless of seed.", + ), + ], + outputs=[ + IO.Image.Output(), + ], + hidden=[ + IO.Hidden.auth_token_comfy_org, + IO.Hidden.api_key_comfy_org, + IO.Hidden.unique_id, + ], + is_api_node=True, + price_badge=IO.PriceBadge( + depends_on=IO.PriceBadgeDepends( + widgets=["model", "model.resolution", "model.number_of_images"], + ), + expr=""" + ( + $isQualityModel := widgets.model = "grok-imagine-image-quality"; + $isPro := $contains(widgets.model, "pro"); + $res := $lookup(widgets, "model.resolution"); + $n := $lookup(widgets, "model.number_of_images"); + $rate := $isQualityModel + ? ($res = "1k" ? 0.05 : 0.07) + : ($isPro ? 0.07 : 0.02); + $base := $isQualityModel ? 0.01 : 0.002; + $output := $rate * $n; + $isPro + ? {"type":"usd","usd": $base + $output} + : {"type":"range_usd","min_usd": $base + $output, "max_usd": 3 * $base + $output} + ) + """, + ), + ) + + @classmethod + async def execute( + cls, + prompt: str, + model: dict, + seed: int, + ) -> IO.NodeOutput: + validate_string(prompt, strip_whitespace=True, min_length=1) + model_id = model["model"] + resolution = model["resolution"] + number_of_images = model["number_of_images"] + images_dict = model.get("images") or {} + aspect_ratio = model.get("aspect_ratio", "auto") + + image_tensors: list[Input.Image] = [t for t in images_dict.values() if t is not None] + n_images = sum(get_number_of_images(t) for t in image_tensors) + if n_images < 1: + raise ValueError("At least one image is required for editing.") + if model_id == "grok-imagine-image-pro" and n_images > 1: + raise ValueError("The pro model supports only 1 input image.") + if model_id != "grok-imagine-image-pro" and n_images > 3: + raise ValueError("A maximum of 3 input images is supported.") + if aspect_ratio != "auto" and n_images == 1: + raise ValueError( + "Custom aspect ratio is only allowed when multiple images are connected to the image input." + ) + + flat_tensors: list[torch.Tensor] = [] + for tensor in image_tensors: + if len(tensor.shape) == 4: + flat_tensors.extend(tensor[i] for i in range(tensor.shape[0])) + else: + flat_tensors.append(tensor) + + response = await sync_op( + cls, + ApiEndpoint(path="/proxy/xai/v1/images/edits", method="POST"), + data=ImageEditRequest( + model=model_id, + images=[ + InputUrlObject(url=f"data:image/png;base64,{tensor_to_base64_string(i)}") for i in flat_tensors + ], + prompt=prompt, + resolution=resolution.lower(), + n=number_of_images, + seed=seed, + aspect_ratio=None if aspect_ratio == "auto" else aspect_ratio, + ), + response_model=ImageGenerationResponse, + price_extractor=_extract_grok_price, + ) + if len(response.data) == 1: + return IO.NodeOutput(await download_url_to_image_tensor(response.data[0].url)) + return IO.NodeOutput( + torch.cat( + [await download_url_to_image_tensor(i) for i in [str(d.url) for d in response.data if d.url]], + ) + ) + + class GrokVideoNode(IO.ComfyNode): @classmethod @@ -737,6 +930,7 @@ class GrokExtension(ComfyExtension): return [ GrokImageNode, GrokImageEditNode, + GrokImageEditNodeV2, GrokVideoNode, GrokVideoReferenceNode, GrokVideoEditNode, diff --git a/comfy_api_nodes/nodes_openai.py b/comfy_api_nodes/nodes_openai.py index daed495da..a5a188634 100644 --- a/comfy_api_nodes/nodes_openai.py +++ b/comfy_api_nodes/nodes_openai.py @@ -27,6 +27,7 @@ from comfy_api_nodes.util import ( ApiEndpoint, download_url_to_bytesio, downscale_image_tensor, + get_number_of_images, poll_op, sync_op, tensor_to_base64_string, @@ -372,6 +373,7 @@ class OpenAIGPTImage1(IO.ComfyNode): display_name="OpenAI GPT Image 2", category="api node/image/OpenAI", description="Generates images synchronously via OpenAI's GPT Image endpoint.", + is_deprecated=True, inputs=[ IO.String.Input( "prompt", @@ -640,6 +642,316 @@ class OpenAIGPTImage1(IO.ComfyNode): return IO.NodeOutput(await validate_and_cast_response(response)) +def _gpt_image_shared_inputs(): + """Inputs shared by all GPT Image models (quality + reference images + mask).""" + return [ + IO.Combo.Input( + "quality", + default="low", + options=["low", "medium", "high"], + tooltip="Image quality, affects cost and generation time.", + ), + IO.Autogrow.Input( + "images", + template=IO.Autogrow.TemplateNames( + IO.Image.Input("image"), + names=[f"image_{i}" for i in range(1, 17)], + min=0, + ), + tooltip="Optional reference image(s) for image editing. Up to 16 images.", + ), + IO.Mask.Input( + "mask", + optional=True, + tooltip="Optional mask for inpainting (white areas will be replaced). " + "Requires exactly one reference image.", + ), + ] + + +def _gpt_image_legacy_model_inputs(): + """Per-model widget set for legacy gpt-image-1 / gpt-image-1.5 (4 base sizes, transparent bg allowed).""" + return [ + IO.Combo.Input( + "size", + default="auto", + options=["auto", "1024x1024", "1024x1536", "1536x1024"], + tooltip="Image size.", + ), + IO.Combo.Input( + "background", + default="auto", + options=["auto", "opaque", "transparent"], + tooltip="Return image with or without background.", + ), + *_gpt_image_shared_inputs(), + ] + + +class OpenAIGPTImageNodeV2(IO.ComfyNode): + + @classmethod + def define_schema(cls): + return IO.Schema( + node_id="OpenAIGPTImageNodeV2", + display_name="OpenAI GPT Image 2", + category="api node/image/OpenAI", + description="Generates images via OpenAI's GPT Image endpoint.", + inputs=[ + IO.String.Input( + "prompt", + default="", + multiline=True, + tooltip="Text prompt for GPT Image", + ), + IO.DynamicCombo.Input( + "model", + options=[ + IO.DynamicCombo.Option( + "gpt-image-2", + [ + IO.Combo.Input( + "size", + default="auto", + options=[ + "auto", + "1024x1024", + "1024x1536", + "1536x1024", + "2048x2048", + "2048x1152", + "1152x2048", + "3840x2160", + "2160x3840", + "Custom", + ], + tooltip="Image size. Select 'Custom' to use the custom width and height.", + ), + IO.Int.Input( + "custom_width", + default=1024, + min=1024, + max=3840, + step=16, + tooltip="Used only when `size` is 'Custom'. Must be a multiple of 16.", + ), + IO.Int.Input( + "custom_height", + default=1024, + min=1024, + max=3840, + step=16, + tooltip="Used only when `size` is 'Custom'. Must be a multiple of 16.", + ), + IO.Combo.Input( + "background", + default="auto", + options=["auto", "opaque"], + tooltip="Return image with or without background.", + ), + *_gpt_image_shared_inputs(), + ], + ), + IO.DynamicCombo.Option("gpt-image-1.5", _gpt_image_legacy_model_inputs()), + IO.DynamicCombo.Option("gpt-image-1", _gpt_image_legacy_model_inputs()), + ], + ), + IO.Int.Input( + "n", + default=1, + min=1, + max=8, + step=1, + tooltip="How many images to generate", + display_mode=IO.NumberDisplay.number, + ), + IO.Int.Input( + "seed", + default=0, + min=0, + max=2147483647, + step=1, + display_mode=IO.NumberDisplay.number, + control_after_generate=True, + tooltip="not implemented yet in backend", + ), + ], + outputs=[IO.Image.Output()], + hidden=[ + IO.Hidden.auth_token_comfy_org, + IO.Hidden.api_key_comfy_org, + IO.Hidden.unique_id, + ], + is_api_node=True, + price_badge=IO.PriceBadge( + depends_on=IO.PriceBadgeDepends(widgets=["model", "model.quality", "n"]), + expr=""" + ( + $ranges := { + "gpt-image-1": { + "low": [0.011, 0.02], + "medium": [0.042, 0.07], + "high": [0.167, 0.25] + }, + "gpt-image-1.5": { + "low": [0.009, 0.02], + "medium": [0.034, 0.062], + "high": [0.133, 0.22] + }, + "gpt-image-2": { + "low": [0.0048, 0.019], + "medium": [0.041, 0.168], + "high": [0.165, 0.67] + } + }; + $range := $lookup($lookup($ranges, widgets.model), $lookup(widgets, "model.quality")); + $nRaw := widgets.n; + $n := ($nRaw != null and $nRaw != 0) ? $nRaw : 1; + ($n = 1) + ? {"type":"range_usd","min_usd": $range[0], "max_usd": $range[1], "format": {"approximate": true}} + : { + "type":"range_usd", + "min_usd": $range[0] * $n, + "max_usd": $range[1] * $n, + "format": { "suffix": "/Run", "approximate": true } + } + ) + """, + ), + ) + + @classmethod + async def execute( + cls, + prompt: str, + model: dict, + n: int, + seed: int, + ) -> IO.NodeOutput: + validate_string(prompt, strip_whitespace=False) + + model_id = model["model"] + size = model["size"] + background = model["background"] + quality = model["quality"] + custom_width = model.get("custom_width", 1024) + custom_height = model.get("custom_height", 1024) + + images_dict = model.get("images") or {} + image_tensors: list[Input.Image] = [t for t in images_dict.values() if t is not None] + n_images = sum(get_number_of_images(t) for t in image_tensors) + mask = model.get("mask") + + if mask is not None and n_images == 0: + raise ValueError("Cannot use a mask without an input image") + + if size == "Custom": + if custom_width % 16 != 0 or custom_height % 16 != 0: + raise ValueError( + f"Custom width and height must be multiples of 16, got {custom_width}x{custom_height}" + ) + if max(custom_width, custom_height) > 3840: + raise ValueError( + f"Custom resolution max edge must be <= 3840, got {custom_width}x{custom_height}" + ) + ratio = max(custom_width, custom_height) / min(custom_width, custom_height) + if ratio > 3: + raise ValueError( + f"Custom resolution aspect ratio must not exceed 3:1, got {custom_width}x{custom_height}" + ) + total_pixels = custom_width * custom_height + if not 655_360 <= total_pixels <= 8_294_400: + raise ValueError( + f"Custom resolution total pixels must be between 655,360 and 8,294,400, got {total_pixels}" + ) + size = f"{custom_width}x{custom_height}" + + if model_id == "gpt-image-1": + price_extractor = calculate_tokens_price_image_1 + elif model_id == "gpt-image-1.5": + price_extractor = calculate_tokens_price_image_1_5 + elif model_id == "gpt-image-2": + price_extractor = calculate_tokens_price_image_2_0 + else: + raise ValueError(f"Unknown model: {model_id}") + + if image_tensors: + flat: list[torch.Tensor] = [] + for tensor in image_tensors: + if len(tensor.shape) == 4: + flat.extend(tensor[i : i + 1] for i in range(tensor.shape[0])) + else: + flat.append(tensor.unsqueeze(0)) + + files = [] + for i, single_image in enumerate(flat): + scaled_image = downscale_image_tensor(single_image, total_pixels=2048 * 2048).squeeze() + image_np = (scaled_image.numpy() * 255).astype(np.uint8) + img = Image.fromarray(image_np) + img_byte_arr = BytesIO() + img.save(img_byte_arr, format="PNG") + img_byte_arr.seek(0) + + if len(flat) == 1: + files.append(("image", (f"image_{i}.png", img_byte_arr, "image/png"))) + else: + files.append(("image[]", (f"image_{i}.png", img_byte_arr, "image/png"))) + + if mask is not None: + if len(flat) != 1: + raise Exception("Cannot use a mask with multiple image") + ref_image = flat[0] + if mask.shape[1:] != ref_image.shape[1:-1]: + raise Exception("Mask and Image must be the same size") + _, height, width = mask.shape + rgba_mask = torch.zeros(height, width, 4, device="cpu") + rgba_mask[:, :, 3] = 1 - mask.squeeze().cpu() + scaled_mask = downscale_image_tensor( + rgba_mask.unsqueeze(0), total_pixels=2048 * 2048 + ).squeeze() + mask_np = (scaled_mask.numpy() * 255).astype(np.uint8) + mask_img = Image.fromarray(mask_np) + mask_img_byte_arr = BytesIO() + mask_img.save(mask_img_byte_arr, format="PNG") + mask_img_byte_arr.seek(0) + files.append(("mask", ("mask.png", mask_img_byte_arr, "image/png"))) + + response = await sync_op( + cls, + ApiEndpoint(path="/proxy/openai/images/edits", method="POST"), + response_model=OpenAIImageGenerationResponse, + data=OpenAIImageEditRequest( + model=model_id, + prompt=prompt, + quality=quality, + background=background, + n=n, + size=size, + moderation="low", + ), + content_type="multipart/form-data", + files=files, + price_extractor=price_extractor, + ) + else: + response = await sync_op( + cls, + ApiEndpoint(path="/proxy/openai/images/generations", method="POST"), + response_model=OpenAIImageGenerationResponse, + data=OpenAIImageGenerationRequest( + model=model_id, + prompt=prompt, + quality=quality, + background=background, + n=n, + size=size, + moderation="low", + ), + price_extractor=price_extractor, + ) + return IO.NodeOutput(await validate_and_cast_response(response)) + + class OpenAIChatNode(IO.ComfyNode): """ Node to generate text responses from an OpenAI model. @@ -999,6 +1311,7 @@ class OpenAIExtension(ComfyExtension): OpenAIDalle2, OpenAIDalle3, OpenAIGPTImage1, + OpenAIGPTImageNodeV2, OpenAIChatNode, OpenAIInputFiles, OpenAIChatConfig, diff --git a/comfy_api_nodes/nodes_quiver.py b/comfy_api_nodes/nodes_quiver.py index 28862e368..3269c0afe 100644 --- a/comfy_api_nodes/nodes_quiver.py +++ b/comfy_api_nodes/nodes_quiver.py @@ -143,7 +143,7 @@ class QuiverTextToSVGNode(IO.ComfyNode): if reference_images: references = [] for key in reference_images: - url = await upload_image_to_comfyapi(cls, reference_images[key]) + url = await upload_image_to_comfyapi(cls, reference_images[key], mime_type="image/png") references.append(QuiverImageObject(url=url)) if len(references) > 4: raise ValueError("Maximum 4 reference images are allowed.") @@ -252,7 +252,7 @@ class QuiverImageToSVGNode(IO.ComfyNode): model: dict, seed: int, ) -> IO.NodeOutput: - image_url = await upload_image_to_comfyapi(cls, image) + image_url = await upload_image_to_comfyapi(cls, image, mime_type="image/png") response = await sync_op( cls, diff --git a/comfy_extras/nodes_advanced_samplers.py b/comfy_extras/nodes_advanced_samplers.py index 7e8411fa4..567c37be0 100644 --- a/comfy_extras/nodes_advanced_samplers.py +++ b/comfy_extras/nodes_advanced_samplers.py @@ -86,6 +86,37 @@ def sample_euler_pp(model, x, sigmas, extra_args=None, callback=None, disable=No return x +class SamplerLCM(io.ComfyNode): + @classmethod + def define_schema(cls) -> io.Schema: + return io.Schema( + node_id="SamplerLCM", + category="sampling/samplers", + description=("LCM sampler with tunable per-step noise. s_noise is a multiplier on the model's training noise scale"), + inputs=[ + io.Float.Input("s_noise", default=1.0, min=0.0, max=64.0, step=0.01, + tooltip="Per-step noise multiplier at the first step (1.0 = match training)."), + io.Float.Input("s_noise_end", default=1.0, min=0.0, max=64.0, step=0.01, + tooltip="Per-step noise multiplier at the last step. Set equal to s_noise for a constant schedule."), + io.Float.Input("noise_clip_std", default=0.0, min=0.0, max=10.0, step=0.01, + tooltip="Clamp per-step noise to +/- N*std. 0 disables."), + ], + outputs=[io.Sampler.Output()], + ) + + @classmethod + def execute(cls, s_noise, s_noise_end, noise_clip_std) -> io.NodeOutput: + sampler = comfy.samplers.ksampler( + "lcm", + { + "s_noise": float(s_noise), + "s_noise_end": float(s_noise_end), + "noise_clip_std": float(noise_clip_std), + }, + ) + return io.NodeOutput(sampler) + + class SamplerEulerCFGpp(io.ComfyNode): @classmethod def define_schema(cls) -> io.Schema: @@ -114,6 +145,7 @@ class AdvancedSamplersExtension(ComfyExtension): async def get_node_list(self) -> list[type[io.ComfyNode]]: return [ SamplerLCMUpscale, + SamplerLCM, SamplerEulerCFGpp, ] diff --git a/comfy_extras/nodes_hidream_o1.py b/comfy_extras/nodes_hidream_o1.py new file mode 100644 index 000000000..f393745f6 --- /dev/null +++ b/comfy_extras/nodes_hidream_o1.py @@ -0,0 +1,256 @@ +from typing_extensions import override + +import torch + +import comfy.model_management +import comfy.patcher_extension +import node_helpers +from comfy_api.latest import ComfyExtension, io + + +class EmptyHiDreamO1LatentImage(io.ComfyNode): + @classmethod + def define_schema(cls) -> io.Schema: + return io.Schema( + node_id="EmptyHiDreamO1LatentImage", + display_name="Empty HiDream-O1 Latent Image", + category="latent/image", + description=( + "Empty pixel-space latent for HiDream-O1-Image. The model was " + "trained at ~4 megapixels; lower resolutions go off-distribution " + "and quality regresses noticeably. Trained resolutions: " + "2048x2048, 2304x1728, 1728x2304, 2560x1440, 1440x2560, " + "2496x1664, 1664x2496, 3104x1312, 1312x3104, 2304x1792, 1792x2304." + ), + inputs=[ + io.Int.Input(id="width", default=2048, min=64, max=4096, step=32), + io.Int.Input(id="height", default=2048, min=64, max=4096, step=32), + io.Int.Input(id="batch_size", default=1, min=1, max=64), + ], + outputs=[io.Latent().Output()], + ) + + @classmethod + def execute(cls, *, width: int, height: int, batch_size: int = 1) -> io.NodeOutput: + latent = torch.zeros( + (batch_size, 3, height, width), + device=comfy.model_management.intermediate_device(), + ) + return io.NodeOutput({"samples": latent}) + + +class HiDreamO1ReferenceImages(io.ComfyNode): + """Attach reference images to both positive and negative conditioning.""" + + @classmethod + def define_schema(cls) -> io.Schema: + return io.Schema( + node_id="HiDreamO1ReferenceImages", + display_name="HiDream-O1 Reference Images", + category="conditioning/image", + description=( + "Attach 1-10 reference images to conditioning, one for edit instruction" + "or multiple for subject-driven personalization." + ), + inputs=[ + io.Conditioning.Input(id="positive"), + io.Conditioning.Input(id="negative"), + io.Autogrow.Input( + "images", + template=io.Autogrow.TemplateNames( + io.Image.Input("image"), + names=[f"image_{i}" for i in range(1, 11)], + min=1, + ), + tooltip=("Reference images. 1 image = instruction edit; 2-10 images = multi reference." + ), + ), + ], + outputs=[ + io.Conditioning.Output(display_name="positive"), + io.Conditioning.Output(display_name="negative"), + ], + ) + + @classmethod + def execute(cls, *, positive, negative, images: io.Autogrow.Type) -> io.NodeOutput: + refs = [images[f"image_{i}"] for i in range(1, 11) if f"image_{i}" in images] + positive = node_helpers.conditioning_set_values(positive, {"reference_latents": refs}, append=True) + negative = node_helpers.conditioning_set_values(negative, {"reference_latents": refs}, append=True) + return io.NodeOutput(positive, negative) + + +class HiDreamO1PatchSeamSmoothing(io.ComfyNode): + PATCH_SIZE = 32 + EDGE_FEATHER = 4 + + # Shift presets per (pattern, N). 8-pass = 4-quadrant + 4 quarter-patch offsets. + SHIFTS_BY_PATTERN = { + ("single_shift", 2): [(0, 0), (16, 16)], + ("single_shift", 4): [(0, 0), (16, 0), (0, 16), (16, 16)], + ("single_shift", 8): [(0, 0), (16, 0), (0, 16), (16, 16), + (8, 8), (24, 8), (8, 24), (24, 24)], + ("symmetric", 2): [(-8, -8), (8, 8)], + ("symmetric", 4): [(-8, -8), (8, -8), (-8, 8), (8, 8)], + ("symmetric", 8): [(-12, -12), (4, -12), (-12, 4), (4, 4), + (-4, -4), (12, -4), (-4, 12), (12, 12)], + } + RAMP_LEVELS = { + "2": [2], + "4": [4], + "ramp_2_4": [2, 4], + "ramp_2_4_8": [2, 4, 8], + } + + @staticmethod + def _hann_tile(cy: int, cx: int, size: int = 32) -> torch.Tensor: + """size x size Hann tile peaking at (cy, cx) within a patch.""" + half = size // 2 + yy = torch.arange(size).view(size, 1) + xx = torch.arange(size).view(1, size) + dy = ((yy - cy + half) % size) - half + dx = ((xx - cx + half) % size) - half + return 0.25 * (1 + torch.cos(torch.pi * dy / half)) * (1 + torch.cos(torch.pi * dx / half)) + + @classmethod + def define_schema(cls) -> io.Schema: + return io.Schema( + node_id="HiDreamO1PatchSeamSmoothing", + display_name="HiDream-O1 Patch Seam Smoothing", + category="advanced/model", + is_experimental=True, + description=( + "Average the model output across multiple shifted patch-grid " + "positions during the late portion of sampling. Cancels seams." + ), + inputs=[ + io.Model.Input(id="model"), + io.Float.Input(id="start_percent", default=0.8, min=0.0, max=1.0, step=0.01, + tooltip="Sampling progress (0=start, 1=end) at which the blend turns ON.", + ), + io.Float.Input(id="end_percent", default=1.0, min=0.0, max=1.0, step=0.01, + tooltip="Sampling progress at which the blend turns OFF.", + ), + io.Combo.Input( + id="pattern", + options=["single_shift", "symmetric"], + default="single_shift", + tooltip="Shift layout. single_shift: one pass at the natural patch grid + others offset. symmetric: all passes off-grid, shifts split around origin.", + ), + io.Combo.Input( + id="passes", + options=["2", "4", "ramp_2_4", "ramp_2_4_8"], + default="2", + tooltip="Number of passes per gated step. 2/4 = fixed. ramp_*: pass count increases as sampling approaches end (more smoothing where seams are most visible).", + ), + io.Combo.Input( + id="blend", + options=["average", "window", "median"], + default="average", + tooltip="average: equal-weight mean. window: Hann-windowed weighting favoring each pass away from its patch boundaries. median: per-pixel median, rejects wraparound-outlier passes.", + ), + io.Float.Input(id="strength", default=1.0, min=0.0, max=1.0, step=0.01, + tooltip="Interpolation between the natural-grid pred (0) and the averaged result (1).", + ), + ], + outputs=[io.Model.Output()], + ) + + @classmethod + def execute(cls, *, model, start_percent: float, end_percent: float, pattern: str, passes: str, blend: str, strength: float) -> io.NodeOutput: + if strength <= 0.0 or end_percent <= start_percent: + return io.NodeOutput(model) + + P = cls.PATCH_SIZE + half = P // 2 + shift_levels = [cls.SHIFTS_BY_PATTERN[(pattern, n)] for n in cls.RAMP_LEVELS[passes]] + + if blend == "window": + window_tile_levels = [ + torch.stack([cls._hann_tile((half - sy) % P, (half - sx) % P, P) for sy, sx in lst], dim=0) + for lst in shift_levels + ] + else: + window_tile_levels = [None] * len(shift_levels) + + m = model.clone() + model_sampling = m.get_model_object("model_sampling") + multiplier = float(model_sampling.multiplier) + start_t = float(model_sampling.percent_to_sigma(start_percent)) * multiplier + end_t = float(model_sampling.percent_to_sigma(end_percent)) * multiplier + + edge_ramp_cache: dict = {} + + def get_edge_ramp(H: int, W: int, device, dtype) -> torch.Tensor: + key = (H, W, device, dtype) + cached = edge_ramp_cache.get(key) + if cached is not None: + return cached + feather = cls.EDGE_FEATHER + ys = torch.minimum(torch.arange(H, device=device, dtype=torch.float32), + (H - 1) - torch.arange(H, device=device, dtype=torch.float32)) + xs = torch.minimum(torch.arange(W, device=device, dtype=torch.float32), + (W - 1) - torch.arange(W, device=device, dtype=torch.float32)) + y_mask = ((ys - P) / feather).clamp(0, 1) + x_mask = ((xs - P) / feather).clamp(0, 1) + ramp = (y_mask[:, None] * x_mask[None, :]).to(dtype) + edge_ramp_cache[key] = ramp + return ramp + + def smoothing_wrapper(executor, *args, **kwargs): + x = args[0] + t = float(args[1][0]) + pred = executor(*args, **kwargs) + if not (end_t <= t <= start_t): + return pred + # Pick shift-level by sigma phase across the gated range. + if len(shift_levels) == 1: + level_idx = 0 + else: + phase = (start_t - t) / max(start_t - end_t, 1e-8) + level_idx = min(int(phase * len(shift_levels)), len(shift_levels) - 1) + shifts = shift_levels[level_idx] + window_tiles = window_tile_levels[level_idx] + + preds = [] + for sy, sx in shifts: + if sy == 0 and sx == 0: + preds.append(pred) + continue + x_rolled = torch.roll(x, shifts=(sy, sx), dims=(-2, -1)) + pred_rolled = executor(x_rolled, *args[1:], **kwargs) + preds.append(torch.roll(pred_rolled, shifts=(-sy, -sx), dims=(-2, -1))) + stacked = torch.stack(preds, dim=0) # (N, B, C, H, W) + _, _, _, H, W = stacked.shape + if blend == "window": + N = stacked.shape[0] + tiles = window_tiles.to(device=stacked.device, dtype=stacked.dtype) + w = tiles.repeat(1, H // P, W // P)[:, :H, :W] + sum_w = w.sum(dim=0, keepdim=True) + w = torch.where(sum_w < 1e-3, torch.full_like(w, 1.0 / N), w / sum_w.clamp(min=1e-8)) + avg = (stacked * w[:, None, None, :, :]).sum(dim=0) + elif blend == "median": + avg = torch.median(stacked, dim=0).values + else: + avg = stacked.mean(dim=0) + + # Mask out the P-px wraparound contamination strip at each edge. + mask = get_edge_ramp(H, W, pred.device, pred.dtype) + return pred * (1.0 - mask * strength) + avg * (mask * strength) + + m.add_wrapper_with_key(comfy.patcher_extension.WrappersMP.DIFFUSION_MODEL, "hidream_o1_patch_seam_smoothing", smoothing_wrapper) + return io.NodeOutput(m) + + +class HiDreamO1Extension(ComfyExtension): + @override + async def get_node_list(self) -> list[type[io.ComfyNode]]: + return [ + EmptyHiDreamO1LatentImage, + HiDreamO1ReferenceImages, + HiDreamO1PatchSeamSmoothing, + ] + + +async def comfy_entrypoint() -> HiDreamO1Extension: + return HiDreamO1Extension() diff --git a/comfy_extras/nodes_model_advanced.py b/comfy_extras/nodes_model_advanced.py index 8bf6a1afa..33b940a0f 100644 --- a/comfy_extras/nodes_model_advanced.py +++ b/comfy_extras/nodes_model_advanced.py @@ -300,6 +300,29 @@ class RescaleCFG: m.set_model_sampler_cfg_function(rescale_cfg) return (m, ) +class ModelNoiseScale: + @classmethod + def INPUT_TYPES(s): + return {"required": { "model": ("MODEL",), + "noise_scale": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 64.0, "step": 0.01, + "tooltip": "Absolute training noise scale. For example HiDream-O1 base: 8.0, dev: 7.5."}), + }} + + RETURN_TYPES = ("MODEL",) + FUNCTION = "patch" + + CATEGORY = "advanced/model" + + def patch(self, model, noise_scale): + m = model.clone() + original = m.model.model_sampling + ms = type(original)(m.model.model_config) + ms.set_parameters(shift=original.shift, multiplier=original.multiplier) + ms.set_noise_scale(noise_scale) + m.add_object_patch("model_sampling", ms) + return (m, ) + + class ModelComputeDtype: SEARCH_ALIASES = ["model precision", "change dtype"] @classmethod @@ -327,6 +350,7 @@ NODE_CLASS_MAPPINGS = { "ModelSamplingSD3": ModelSamplingSD3, "ModelSamplingAuraFlow": ModelSamplingAuraFlow, "ModelSamplingFlux": ModelSamplingFlux, + "ModelNoiseScale": ModelNoiseScale, "RescaleCFG": RescaleCFG, "ModelComputeDtype": ModelComputeDtype, } diff --git a/comfy_extras/nodes_qwen.py b/comfy_extras/nodes_qwen.py index 6894367be..fde8fac9a 100644 --- a/comfy_extras/nodes_qwen.py +++ b/comfy_extras/nodes_qwen.py @@ -116,7 +116,7 @@ class EmptyQwenImageLayeredLatentImage(io.ComfyNode): inputs=[ io.Int.Input("width", default=640, min=16, max=nodes.MAX_RESOLUTION, step=16), io.Int.Input("height", default=640, min=16, max=nodes.MAX_RESOLUTION, step=16), - io.Int.Input("layers", default=3, min=0, max=nodes.MAX_RESOLUTION, step=1, advanced=True), + io.Int.Input("layers", default=3, min=0, max=nodes.MAX_RESOLUTION, step=1), io.Int.Input("batch_size", default=1, min=1, max=4096), ], outputs=[ diff --git a/comfyui_version.py b/comfyui_version.py index 53e7156e3..45626792f 100644 --- a/comfyui_version.py +++ b/comfyui_version.py @@ -1,3 +1,3 @@ # This file is automatically generated by the build process when version is # updated in pyproject.toml. -__version__ = "0.20.1" +__version__ = "0.21.0" diff --git a/nodes.py b/nodes.py index 67ed7ec49..2b63f9fbb 100644 --- a/nodes.py +++ b/nodes.py @@ -2435,6 +2435,7 @@ async def init_builtin_extra_nodes(): "nodes_sam3.py", "nodes_void.py", "nodes_wandancer.py", + "nodes_hidream_o1.py", "nodes_save_3d.py", ] diff --git a/pyproject.toml b/pyproject.toml index 633dac517..825b492ed 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -1,6 +1,6 @@ [project] name = "ComfyUI" -version = "0.20.1" +version = "0.21.0" readme = "README.md" license = { file = "LICENSE" } requires-python = ">=3.10"