mirror of
https://github.com/comfyanonymous/ComfyUI.git
synced 2026-05-10 01:02:56 +08:00
Compare commits
17 Commits
7095c947b5
...
3ad8a0fd06
| Author | SHA1 | Date | |
|---|---|---|---|
|
3ad8a0fd06 | ||
|
|
7bbf1e8169 | ||
|
|
8b08bfdcbe | ||
|
|
4e823431cc | ||
|
|
66669b2ded | ||
|
|
65045730a6 | ||
|
|
87878f354f | ||
|
|
de97192962 | ||
|
|
d56a093800 | ||
|
|
8dd41ef82e | ||
|
|
b715186140 | ||
|
|
ae54d7a987 | ||
|
|
f3aebfa2b0 | ||
|
|
b3a066559b | ||
|
|
0b7d56070d | ||
|
|
c3cd2a4e75 | ||
|
|
800bf842a5 |
@ -431,9 +431,10 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Color adjust"
|
||||
"category": "Image Tools/Color adjust",
|
||||
"description": "Adjusts image brightness and contrast using a real-time GPU fragment shader."
|
||||
}
|
||||
]
|
||||
},
|
||||
"extra": {}
|
||||
}
|
||||
}
|
||||
@ -162,7 +162,7 @@
|
||||
},
|
||||
"revision": 0,
|
||||
"config": {},
|
||||
"name": "local-Canny to Image (Z-Image-Turbo)",
|
||||
"name": "Canny to Image (Z-Image-Turbo)",
|
||||
"inputNode": {
|
||||
"id": -10,
|
||||
"bounding": [
|
||||
@ -1553,7 +1553,8 @@
|
||||
"VHS_MetadataImage": true,
|
||||
"VHS_KeepIntermediate": true
|
||||
},
|
||||
"category": "Image generation and editing/Canny to image"
|
||||
"category": "Image generation and editing/Canny to image",
|
||||
"description": "Generates an image from a Canny edge map using Z-Image-Turbo, with text conditioning."
|
||||
}
|
||||
]
|
||||
},
|
||||
@ -1574,4 +1575,4 @@
|
||||
}
|
||||
},
|
||||
"version": 0.4
|
||||
}
|
||||
}
|
||||
@ -192,7 +192,7 @@
|
||||
},
|
||||
"revision": 0,
|
||||
"config": {},
|
||||
"name": "local-Canny to Video (LTX 2.0)",
|
||||
"name": "Canny to Video (LTX 2.0)",
|
||||
"inputNode": {
|
||||
"id": -10,
|
||||
"bounding": [
|
||||
@ -3600,7 +3600,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Video generation and editing/Canny to video"
|
||||
"category": "Video generation and editing/Canny to video",
|
||||
"description": "Generates video from Canny edge maps using LTX-2, with optional synchronized audio."
|
||||
}
|
||||
]
|
||||
},
|
||||
@ -3616,4 +3617,4 @@
|
||||
}
|
||||
},
|
||||
"version": 0.4
|
||||
}
|
||||
}
|
||||
@ -377,8 +377,9 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Color adjust"
|
||||
"category": "Image Tools/Color adjust",
|
||||
"description": "Adds lens-style chromatic aberration (color fringing) using a real-time GPU fragment shader."
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -596,7 +596,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Color adjust"
|
||||
"category": "Image Tools/Color adjust",
|
||||
"description": "Adjusts saturation, temperature, tint, and vibrance using a real-time GPU fragment shader."
|
||||
}
|
||||
]
|
||||
}
|
||||
|
||||
@ -1129,7 +1129,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Color adjust"
|
||||
"category": "Image Tools/Color adjust",
|
||||
"description": "Balances colors across shadows, midtones, and highlights using a real-time GPU fragment shader."
|
||||
}
|
||||
]
|
||||
}
|
||||
|
||||
@ -608,7 +608,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Color adjust"
|
||||
"category": "Image Tools/Color adjust",
|
||||
"description": "Fine-tunes tone and color with per-channel curve adjustments using a real-time GPU fragment shader."
|
||||
}
|
||||
]
|
||||
}
|
||||
|
||||
@ -1609,7 +1609,8 @@
|
||||
}
|
||||
],
|
||||
"extra": {},
|
||||
"category": "Image Tools/Crop"
|
||||
"category": "Image Tools/Crop",
|
||||
"description": "Splits an image into a 2×2 grid of four equal tiles."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -2946,7 +2946,8 @@
|
||||
}
|
||||
],
|
||||
"extra": {},
|
||||
"category": "Image Tools/Crop"
|
||||
"category": "Image Tools/Crop",
|
||||
"description": "Splits an image into a 3×3 grid of nine equal tiles."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -1579,7 +1579,8 @@
|
||||
"VHS_MetadataImage": true,
|
||||
"VHS_KeepIntermediate": true
|
||||
},
|
||||
"category": "Image generation and editing/Depth to image"
|
||||
"category": "Image generation and editing/Depth to image",
|
||||
"description": "Generates an image from a depth map using Z-Image-Turbo with text conditioning."
|
||||
},
|
||||
{
|
||||
"id": "458bdf3c-4b58-421c-af50-c9c663a4d74c",
|
||||
@ -2461,7 +2462,8 @@
|
||||
]
|
||||
},
|
||||
"workflowRendererVersion": "LG"
|
||||
}
|
||||
},
|
||||
"description": "Estimates a monocular depth map from an input image using the Lotus depth estimation model."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -4233,7 +4233,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Video generation and editing/Depth to video"
|
||||
"category": "Video generation and editing/Depth to video",
|
||||
"description": "Generates video from depth maps using LTX-2, with optional synchronized audio."
|
||||
},
|
||||
{
|
||||
"id": "38b60539-50a7-42f9-a5fe-bdeca26272e2",
|
||||
@ -5192,7 +5193,8 @@
|
||||
],
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
}
|
||||
},
|
||||
"description": "Estimates a monocular depth map from an input image using the Lotus depth estimation model."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -450,9 +450,10 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Blur"
|
||||
"category": "Image Tools/Blur",
|
||||
"description": "Applies bilateral (edge-preserving) blur to soften images while retaining detail."
|
||||
}
|
||||
]
|
||||
},
|
||||
"extra": {}
|
||||
}
|
||||
}
|
||||
@ -580,8 +580,9 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Color adjust"
|
||||
"category": "Image Tools/Color adjust",
|
||||
"description": "Adds procedural film grain texture for a cinematic look via GPU fragment shader."
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -3350,7 +3350,8 @@
|
||||
}
|
||||
],
|
||||
"extra": {},
|
||||
"category": "Video generation and editing/First-Last-Frame to Video"
|
||||
"category": "Video generation and editing/First-Last-Frame to Video",
|
||||
"description": "Generates a video interpolating between first and last keyframes using LTX-2.3."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -575,8 +575,9 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Color adjust"
|
||||
"category": "Image Tools/Color adjust",
|
||||
"description": "Adds a glow/bloom effect around bright image areas via GPU fragment shader."
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -752,8 +752,9 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Color adjust"
|
||||
"category": "Image Tools/Color adjust",
|
||||
"description": "Adjusts hue, saturation, and lightness of an image using a real-time GPU fragment shader."
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -374,7 +374,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Blur"
|
||||
"category": "Image Tools/Blur",
|
||||
"description": "Applies Gaussian, Box, or Radial blur to soften images and create stylized depth or motion effects."
|
||||
}
|
||||
]
|
||||
}
|
||||
|
||||
@ -310,7 +310,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Text generation/Image Captioning"
|
||||
"category": "Text generation/Image Captioning",
|
||||
"description": "Generates descriptive captions for images using Google's Gemini multimodal LLM."
|
||||
}
|
||||
]
|
||||
}
|
||||
|
||||
@ -315,8 +315,9 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Color adjust"
|
||||
"category": "Image Tools/Color adjust",
|
||||
"description": "Manipulates individual RGBA channels for masking, compositing, and channel effects."
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -2138,7 +2138,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Edit image"
|
||||
"category": "Image generation and editing/Edit image",
|
||||
"description": "Edits images via text instructions using FireRed Image Edit 1.1, a diffusion-based instruction-following editing model."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -1472,7 +1472,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Edit image"
|
||||
"category": "Image generation and editing/Edit image",
|
||||
"description": "Edits an input image via text instructions using FLUX.2 [klein] 4B."
|
||||
},
|
||||
{
|
||||
"id": "6007e698-2ebd-4917-84d8-299b35d7b7ab",
|
||||
@ -1821,7 +1822,8 @@
|
||||
],
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
}
|
||||
},
|
||||
"description": "Applies reference image conditioning for style/identity transfer (Flux.2 Klein 4B)."
|
||||
}
|
||||
]
|
||||
},
|
||||
@ -1837,4 +1839,4 @@
|
||||
}
|
||||
},
|
||||
"version": 0.4
|
||||
}
|
||||
}
|
||||
|
||||
@ -1417,7 +1417,8 @@
|
||||
}
|
||||
],
|
||||
"extra": {},
|
||||
"category": "Image generation and editing/Edit image"
|
||||
"category": "Image generation and editing/Edit image",
|
||||
"description": "Edits images via text instructions using LongCat Image Edit, an instruction-following image editing diffusion model."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -132,7 +132,7 @@
|
||||
},
|
||||
"revision": 0,
|
||||
"config": {},
|
||||
"name": "local-Image Edit (Qwen 2511)",
|
||||
"name": "Image Edit (Qwen 2511)",
|
||||
"inputNode": {
|
||||
"id": -10,
|
||||
"bounding": [
|
||||
@ -1468,7 +1468,8 @@
|
||||
"VHS_MetadataImage": true,
|
||||
"VHS_KeepIntermediate": true
|
||||
},
|
||||
"category": "Image generation and editing/Edit image"
|
||||
"category": "Image generation and editing/Edit image",
|
||||
"description": "Edits images via text instructions using Qwen-Image-Edit-2511 with improved character consistency and integrated LoRA."
|
||||
}
|
||||
]
|
||||
},
|
||||
@ -1489,4 +1490,4 @@
|
||||
}
|
||||
},
|
||||
"version": 0.4
|
||||
}
|
||||
}
|
||||
@ -1188,7 +1188,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Inpaint image"
|
||||
"category": "Image generation and editing/Inpaint image",
|
||||
"description": "Inpaints masked image regions using Flux.1 fill [dev], Black Forest Labs' inpainting/outpainting model."
|
||||
}
|
||||
]
|
||||
},
|
||||
@ -1202,4 +1203,4 @@
|
||||
},
|
||||
"ue_links": []
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@ -1548,7 +1548,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Inpaint image"
|
||||
"category": "Image generation and editing/Inpaint image",
|
||||
"description": "Inpaints masked regions using Qwen-Image, extending its multilingual text rendering to inpainting tasks."
|
||||
},
|
||||
{
|
||||
"id": "56a1f603-fbd2-40ed-94ef-c9ecbd96aca8",
|
||||
@ -1907,7 +1908,8 @@
|
||||
],
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
}
|
||||
},
|
||||
"description": "Expands and softens mask edges to reduce visible seams after image processing."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -742,9 +742,10 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Color adjust"
|
||||
"category": "Image Tools/Color adjust",
|
||||
"description": "Adjusts black point, white point, and gamma for tonal range control via GPU shader."
|
||||
}
|
||||
]
|
||||
},
|
||||
"extra": {}
|
||||
}
|
||||
}
|
||||
@ -1919,7 +1919,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Outpaint image"
|
||||
"category": "Image generation and editing/Outpaint image",
|
||||
"description": "Outpaints beyond image boundaries using Qwen-Image's outpainting capabilities."
|
||||
},
|
||||
{
|
||||
"id": "f93c215e-c393-460e-9534-ed2c3d8a652e",
|
||||
@ -2278,7 +2279,8 @@
|
||||
],
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
}
|
||||
},
|
||||
"description": "Expands and softens mask edges to reduce visible seams after image processing."
|
||||
},
|
||||
{
|
||||
"id": "2a4b2cc0-db37-4302-a067-da392f38f06b",
|
||||
@ -2733,7 +2735,8 @@
|
||||
],
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
}
|
||||
},
|
||||
"description": "Scales both image and mask together while preserving alignment for editing workflows."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -141,7 +141,7 @@
|
||||
},
|
||||
"revision": 0,
|
||||
"config": {},
|
||||
"name": "local-Image Upscale(Z-image-Turbo)",
|
||||
"name": "Image Upscale (Z-image-Turbo)",
|
||||
"inputNode": {
|
||||
"id": -10,
|
||||
"bounding": [
|
||||
@ -1302,7 +1302,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Enhance"
|
||||
"category": "Image generation and editing/Enhance",
|
||||
"description": "Upscales images to higher resolution using Z-Image-Turbo."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -99,7 +99,7 @@
|
||||
},
|
||||
"revision": 0,
|
||||
"config": {},
|
||||
"name": "local-Image to Depth Map (Lotus)",
|
||||
"name": "Image to Depth Map (Lotus)",
|
||||
"inputNode": {
|
||||
"id": -10,
|
||||
"bounding": [
|
||||
@ -948,7 +948,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Depth to image"
|
||||
"category": "Image generation and editing/Depth to image",
|
||||
"description": "Estimates a monocular depth map from an input image using the Lotus depth estimation model."
|
||||
}
|
||||
]
|
||||
},
|
||||
@ -964,4 +965,4 @@
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"version": 0.4
|
||||
}
|
||||
}
|
||||
@ -1586,7 +1586,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Image to layers"
|
||||
"category": "Image generation and editing/Image to layers",
|
||||
"description": "Decomposes an image into variable-resolution RGBA layers for independent editing using Qwen-Image-Layered."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -72,7 +72,7 @@
|
||||
},
|
||||
"revision": 0,
|
||||
"config": {},
|
||||
"name": "local-Image to Model (Hunyuan3d 2.1)",
|
||||
"name": "Image to 3D Model (Hunyuan3d 2.1)",
|
||||
"inputNode": {
|
||||
"id": -10,
|
||||
"bounding": [
|
||||
@ -765,7 +765,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "3D/Image to 3D Model"
|
||||
"category": "3D/Image to 3D Model",
|
||||
"description": "Generates 3D mesh models from a single input image using Hunyuan3D 2.0/2.1."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -4223,7 +4223,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "Vue-corrected"
|
||||
},
|
||||
"category": "Video generation and editing/Image to video"
|
||||
"category": "Video generation and editing/Image to video",
|
||||
"description": "Generates video from a single input image using LTX-2.3."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -206,7 +206,7 @@
|
||||
},
|
||||
"revision": 0,
|
||||
"config": {},
|
||||
"name": "local-Image to Video (Wan 2.2)",
|
||||
"name": "Image to Video (Wan 2.2)",
|
||||
"inputNode": {
|
||||
"id": -10,
|
||||
"bounding": [
|
||||
@ -2027,7 +2027,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Video generation and editing/Image to video"
|
||||
"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."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -134,7 +134,7 @@
|
||||
},
|
||||
"revision": 0,
|
||||
"config": {},
|
||||
"name": "local-Pose to Image (Z-Image-Turbo)",
|
||||
"name": "Pose to Image (Z-Image-Turbo)",
|
||||
"inputNode": {
|
||||
"id": -10,
|
||||
"bounding": [
|
||||
@ -1298,7 +1298,8 @@
|
||||
"VHS_MetadataImage": true,
|
||||
"VHS_KeepIntermediate": true
|
||||
},
|
||||
"category": "Image generation and editing/Pose to image"
|
||||
"category": "Image generation and editing/Pose to image",
|
||||
"description": "Generates an image from pose keypoints using Z-Image-Turbo with text conditioning."
|
||||
}
|
||||
]
|
||||
},
|
||||
@ -1319,4 +1320,4 @@
|
||||
}
|
||||
},
|
||||
"version": 0.4
|
||||
}
|
||||
}
|
||||
@ -3870,7 +3870,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Video generation and editing/Pose to video"
|
||||
"category": "Video generation and editing/Pose to video",
|
||||
"description": "Generates video from pose reference frames using LTX-2, with optional synchronized audio."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -270,9 +270,10 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Text generation/Prompt enhance"
|
||||
"category": "Text generation/Prompt enhance",
|
||||
"description": "Expands short text prompts into detailed descriptions using a text generation model for better generation quality."
|
||||
}
|
||||
]
|
||||
},
|
||||
"extra": {}
|
||||
}
|
||||
}
|
||||
@ -302,8 +302,9 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Sharpen"
|
||||
"category": "Image Tools/Sharpen",
|
||||
"description": "Sharpens image details using a GPU fragment shader for enhanced clarity."
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -222,7 +222,7 @@
|
||||
},
|
||||
"revision": 0,
|
||||
"config": {},
|
||||
"name": "local-Text to Audio (ACE-Step 1.5)",
|
||||
"name": "Text to Audio (ACE-Step 1.5)",
|
||||
"inputNode": {
|
||||
"id": -10,
|
||||
"bounding": [
|
||||
@ -1502,7 +1502,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Audio/Music generation"
|
||||
"category": "Audio/Music generation",
|
||||
"description": "Generates audio/music from text prompts using ACE-Step 1.5, a diffusion-based audio generation model."
|
||||
}
|
||||
]
|
||||
},
|
||||
@ -1518,4 +1519,4 @@
|
||||
}
|
||||
},
|
||||
"version": 0.4
|
||||
}
|
||||
}
|
||||
@ -1029,7 +1029,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Text to image"
|
||||
"category": "Image generation and editing/Text to image",
|
||||
"description": "Generates images from text prompts using Flux.1 [dev], Black Forest Labs' 12B diffusion model."
|
||||
}
|
||||
]
|
||||
},
|
||||
@ -1043,4 +1044,4 @@
|
||||
},
|
||||
"ue_links": []
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@ -1023,7 +1023,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Text to image"
|
||||
"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."
|
||||
}
|
||||
]
|
||||
},
|
||||
@ -1037,4 +1038,4 @@
|
||||
},
|
||||
"ue_links": []
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@ -1104,7 +1104,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Text to image"
|
||||
"category": "Image generation and editing/Text to image",
|
||||
"description": "Generates images from text prompts using NetaYume Lumina, fine-tuned from Neta Lumina for anime-style and illustration generation."
|
||||
},
|
||||
{
|
||||
"id": "a07fdf06-1bda-4dac-bdbd-63ee8ebca1c9",
|
||||
@ -1458,11 +1459,12 @@
|
||||
],
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
}
|
||||
},
|
||||
"description": "Encodes a negative text prompt via CLIP for classifier-free guidance in anime-style generation (NetaYume Lumina)."
|
||||
}
|
||||
]
|
||||
},
|
||||
"extra": {
|
||||
"ue_links": []
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@ -1941,7 +1941,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "Vue-corrected"
|
||||
},
|
||||
"category": "Image generation and editing/Text to image"
|
||||
"category": "Image generation and editing/Text to image",
|
||||
"description": "Generates images from text prompts using Qwen-Image-2512, with enhanced human realism and finer natural detail over the base version."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -1873,7 +1873,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Text to image"
|
||||
"category": "Image generation and editing/Text to image",
|
||||
"description": "Generates images from text prompts using Qwen-Image, Alibaba's 20B MMDiT model with excellent multilingual text rendering."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -149,7 +149,7 @@
|
||||
},
|
||||
"revision": 0,
|
||||
"config": {},
|
||||
"name": "local-Text to Image (Z-Image-Turbo)",
|
||||
"name": "Text to Image (Z-Image-Turbo)",
|
||||
"inputNode": {
|
||||
"id": -10,
|
||||
"bounding": [
|
||||
@ -1054,7 +1054,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image generation and editing/Text to image"
|
||||
"category": "Image generation and editing/Text to image",
|
||||
"description": "Generates images from text prompts using Z-Image-Turbo, Alibaba's distilled 6B DiT model."
|
||||
}
|
||||
]
|
||||
},
|
||||
@ -1075,4 +1076,4 @@
|
||||
}
|
||||
},
|
||||
"version": 0.4
|
||||
}
|
||||
}
|
||||
@ -4286,7 +4286,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "Vue-corrected"
|
||||
},
|
||||
"category": "Video generation and editing/Text to video"
|
||||
"category": "Video generation and editing/Text to video",
|
||||
"description": "Generates video from text prompts using LTX-2.3, Lightricks' video diffusion model."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -1572,7 +1572,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Video generation and editing/Text to video"
|
||||
"category": "Video generation and editing/Text to video",
|
||||
"description": "Generates video from text prompts using Wan2.2, Alibaba's diffusion video model."
|
||||
}
|
||||
]
|
||||
},
|
||||
@ -1586,4 +1587,4 @@
|
||||
"VHS_KeepIntermediate": true
|
||||
},
|
||||
"version": 0.4
|
||||
}
|
||||
}
|
||||
@ -434,8 +434,9 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Image Tools/Sharpen"
|
||||
"category": "Image Tools/Sharpen",
|
||||
"description": "Enhances edge contrast via unsharp masking for a sharper image appearance."
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -307,7 +307,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Text generation/Video Captioning"
|
||||
"category": "Text generation/Video Captioning",
|
||||
"description": "Generates descriptive captions for video input using Google's Gemini multimodal LLM."
|
||||
}
|
||||
]
|
||||
}
|
||||
|
||||
@ -165,7 +165,7 @@
|
||||
},
|
||||
"revision": 0,
|
||||
"config": {},
|
||||
"name": "local-Video Inpaint(Wan2.1 VACE)",
|
||||
"name": "Video Inpaint (Wan 2.1 VACE)",
|
||||
"inputNode": {
|
||||
"id": -10,
|
||||
"bounding": [
|
||||
@ -2368,7 +2368,8 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Video generation and editing/Inpaint video"
|
||||
"category": "Video generation and editing/Inpaint video",
|
||||
"description": "Inpaints masked regions in video frames using Wan 2.1 VACE."
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
@ -584,8 +584,9 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Video Tools/Stitch videos"
|
||||
"category": "Video Tools/Stitch videos",
|
||||
"description": "Stitches multiple video clips into a single sequential video file."
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -412,9 +412,10 @@
|
||||
"extra": {
|
||||
"workflowRendererVersion": "LG"
|
||||
},
|
||||
"category": "Video generation and editing/Enhance video"
|
||||
"category": "Video generation and editing/Enhance video",
|
||||
"description": "Upscales video to 4× resolution using a GAN-based upscaling model."
|
||||
}
|
||||
]
|
||||
},
|
||||
"extra": {}
|
||||
}
|
||||
}
|
||||
@ -1390,7 +1390,7 @@ def convert_old_quants(state_dict, model_prefix="", metadata={}):
|
||||
k_out = "{}.weight_scale".format(layer)
|
||||
|
||||
if layer is not None:
|
||||
layer_conf = {"format": "float8_e4m3fn"} # TODO: check if anyone did some non e4m3fn scaled checkpoints
|
||||
layer_conf = {"format": "float8_e4m3fn"}
|
||||
if full_precision_matrix_mult:
|
||||
layer_conf["full_precision_matrix_mult"] = full_precision_matrix_mult
|
||||
layers[layer] = layer_conf
|
||||
|
||||
@ -1,10 +1,11 @@
|
||||
from __future__ import annotations
|
||||
from enum import Enum
|
||||
from typing import Optional, List, Dict, Any, Union
|
||||
from typing import Optional, Any
|
||||
|
||||
from pydantic import BaseModel, Field, RootModel
|
||||
|
||||
|
||||
class TripoModelVersion(str, Enum):
|
||||
v3_1_20260211 = 'v3.1-20260211'
|
||||
v3_0_20250812 = 'v3.0-20250812'
|
||||
v2_5_20250123 = 'v2.5-20250123'
|
||||
v2_0_20240919 = 'v2.0-20240919'
|
||||
@ -142,7 +143,7 @@ class TripoFileEmptyReference(BaseModel):
|
||||
pass
|
||||
|
||||
class TripoFileReference(RootModel):
|
||||
root: Union[TripoFileTokenReference, TripoUrlReference, TripoObjectReference, TripoFileEmptyReference]
|
||||
root: TripoFileTokenReference | TripoUrlReference | TripoObjectReference | TripoFileEmptyReference
|
||||
|
||||
class TripoGetStsTokenRequest(BaseModel):
|
||||
format: str = Field(..., description='The format of the image')
|
||||
@ -183,7 +184,7 @@ class TripoImageToModelRequest(BaseModel):
|
||||
|
||||
class TripoMultiviewToModelRequest(BaseModel):
|
||||
type: TripoTaskType = TripoTaskType.MULTIVIEW_TO_MODEL
|
||||
files: List[TripoFileReference] = Field(..., description='The file references to convert to a model')
|
||||
files: list[TripoFileReference] = Field(..., description='The file references to convert to a model')
|
||||
model_version: Optional[TripoModelVersion] = Field(None, description='The model version to use for generation')
|
||||
orthographic_projection: Optional[bool] = Field(False, description='Whether to use orthographic projection')
|
||||
face_limit: Optional[int] = Field(None, description='The number of faces to limit the generation to')
|
||||
@ -251,27 +252,13 @@ class TripoConvertModelRequest(BaseModel):
|
||||
with_animation: Optional[bool] = Field(None, description='Whether to include animations')
|
||||
pack_uv: Optional[bool] = Field(None, description='Whether to pack the UVs')
|
||||
bake: Optional[bool] = Field(None, description='Whether to bake the model')
|
||||
part_names: Optional[List[str]] = Field(None, description='The names of the parts to include')
|
||||
part_names: Optional[list[str]] = Field(None, description='The names of the parts to include')
|
||||
fbx_preset: Optional[TripoFbxPreset] = Field(None, description='The preset for the FBX export')
|
||||
export_vertex_colors: Optional[bool] = Field(None, description='Whether to export the vertex colors')
|
||||
export_orientation: Optional[TripoOrientation] = Field(None, description='The orientation for the export')
|
||||
animate_in_place: Optional[bool] = Field(None, description='Whether to animate in place')
|
||||
|
||||
|
||||
class TripoTaskRequest(RootModel):
|
||||
root: Union[
|
||||
TripoTextToModelRequest,
|
||||
TripoImageToModelRequest,
|
||||
TripoMultiviewToModelRequest,
|
||||
TripoTextureModelRequest,
|
||||
TripoRefineModelRequest,
|
||||
TripoAnimatePrerigcheckRequest,
|
||||
TripoAnimateRigRequest,
|
||||
TripoAnimateRetargetRequest,
|
||||
TripoStylizeModelRequest,
|
||||
TripoConvertModelRequest
|
||||
]
|
||||
|
||||
class TripoTaskOutput(BaseModel):
|
||||
model: Optional[str] = Field(None, description='URL to the model')
|
||||
base_model: Optional[str] = Field(None, description='URL to the base model')
|
||||
@ -283,12 +270,13 @@ class TripoTask(BaseModel):
|
||||
task_id: str = Field(..., description='The task ID')
|
||||
type: Optional[str] = Field(None, description='The type of task')
|
||||
status: Optional[TripoTaskStatus] = Field(None, description='The status of the task')
|
||||
input: Optional[Dict[str, Any]] = Field(None, description='The input parameters for the task')
|
||||
input: Optional[dict[str, Any]] = Field(None, description='The input parameters for the task')
|
||||
output: Optional[TripoTaskOutput] = Field(None, description='The output of the task')
|
||||
progress: Optional[int] = Field(None, description='The progress of the task', ge=0, le=100)
|
||||
create_time: Optional[int] = Field(None, description='The creation time of the task')
|
||||
running_left_time: Optional[int] = Field(None, description='The estimated time left for the task')
|
||||
queue_position: Optional[int] = Field(None, description='The position in the queue')
|
||||
consumed_credit: int | None = Field(None)
|
||||
|
||||
class TripoTaskResponse(BaseModel):
|
||||
code: int = Field(0, description='The response code')
|
||||
@ -296,7 +284,7 @@ class TripoTaskResponse(BaseModel):
|
||||
|
||||
class TripoGeneralResponse(BaseModel):
|
||||
code: int = Field(0, description='The response code')
|
||||
data: Dict[str, str] = Field(..., description='The task ID data')
|
||||
data: dict[str, str] = Field(..., description='The task ID data')
|
||||
|
||||
class TripoBalanceData(BaseModel):
|
||||
balance: float = Field(..., description='The account balance')
|
||||
|
||||
@ -60,6 +60,7 @@ async def poll_until_finished(
|
||||
],
|
||||
status_extractor=lambda x: x.data.status,
|
||||
progress_extractor=lambda x: x.data.progress,
|
||||
price_extractor=lambda x: x.data.consumed_credit * 0.01 if x.data.consumed_credit else None,
|
||||
estimated_duration=average_duration,
|
||||
)
|
||||
if response_poll.data.status == TripoTaskStatus.SUCCESS:
|
||||
@ -113,7 +114,6 @@ class TripoTextToModelNode(IO.ComfyNode):
|
||||
depends_on=IO.PriceBadgeDepends(
|
||||
widgets=[
|
||||
"model_version",
|
||||
"style",
|
||||
"texture",
|
||||
"pbr",
|
||||
"quad",
|
||||
@ -124,20 +124,17 @@ class TripoTextToModelNode(IO.ComfyNode):
|
||||
expr="""
|
||||
(
|
||||
$isV14 := $contains(widgets.model_version,"v1.4");
|
||||
$style := widgets.style;
|
||||
$hasStyle := ($style != "" and $style != "none");
|
||||
$isV3OrLater := $contains(widgets.model_version,"v3.");
|
||||
$withTexture := widgets.texture or widgets.pbr;
|
||||
$isHdTexture := (widgets.texture_quality = "detailed");
|
||||
$isDetailedGeometry := (widgets.geometry_quality = "detailed");
|
||||
$baseCredits :=
|
||||
$isV14 ? 20 : ($withTexture ? 20 : 10);
|
||||
$credits :=
|
||||
$baseCredits
|
||||
+ ($hasStyle ? 5 : 0)
|
||||
$credits := $isV14 ? 20 : (
|
||||
($withTexture ? 20 : 10)
|
||||
+ (widgets.quad ? 5 : 0)
|
||||
+ ($isHdTexture ? 10 : 0)
|
||||
+ ($isDetailedGeometry ? 20 : 0);
|
||||
{"type":"usd","usd": $round($credits * 0.01, 2)}
|
||||
+ (($isDetailedGeometry and $isV3OrLater) ? 20 : 0)
|
||||
);
|
||||
{"type":"usd","usd": $round($credits * 0.01, 2), "format": {"approximate": true}}
|
||||
)
|
||||
""",
|
||||
),
|
||||
@ -239,7 +236,6 @@ class TripoImageToModelNode(IO.ComfyNode):
|
||||
depends_on=IO.PriceBadgeDepends(
|
||||
widgets=[
|
||||
"model_version",
|
||||
"style",
|
||||
"texture",
|
||||
"pbr",
|
||||
"quad",
|
||||
@ -250,20 +246,17 @@ class TripoImageToModelNode(IO.ComfyNode):
|
||||
expr="""
|
||||
(
|
||||
$isV14 := $contains(widgets.model_version,"v1.4");
|
||||
$style := widgets.style;
|
||||
$hasStyle := ($style != "" and $style != "none");
|
||||
$isV3OrLater := $contains(widgets.model_version,"v3.");
|
||||
$withTexture := widgets.texture or widgets.pbr;
|
||||
$isHdTexture := (widgets.texture_quality = "detailed");
|
||||
$isDetailedGeometry := (widgets.geometry_quality = "detailed");
|
||||
$baseCredits :=
|
||||
$isV14 ? 30 : ($withTexture ? 30 : 20);
|
||||
$credits :=
|
||||
$baseCredits
|
||||
+ ($hasStyle ? 5 : 0)
|
||||
$credits := $isV14 ? 30 : (
|
||||
($withTexture ? 30 : 20)
|
||||
+ (widgets.quad ? 5 : 0)
|
||||
+ ($isHdTexture ? 10 : 0)
|
||||
+ ($isDetailedGeometry ? 20 : 0);
|
||||
{"type":"usd","usd": $round($credits * 0.01, 2)}
|
||||
+ (($isDetailedGeometry and $isV3OrLater) ? 20 : 0)
|
||||
);
|
||||
{"type":"usd","usd": $round($credits * 0.01, 2), "format": {"approximate": true}}
|
||||
)
|
||||
""",
|
||||
),
|
||||
@ -358,7 +351,7 @@ class TripoMultiviewToModelNode(IO.ComfyNode):
|
||||
"texture_alignment", default="original_image", options=["original_image", "geometry"], optional=True, advanced=True
|
||||
),
|
||||
IO.Int.Input("face_limit", default=-1, min=-1, max=500000, optional=True, advanced=True),
|
||||
IO.Boolean.Input("quad", default=False, optional=True, advanced=True),
|
||||
IO.Boolean.Input("quad", default=False, optional=True, advanced=True, tooltip="This parameter is deprecated and does nothing."),
|
||||
IO.Combo.Input("geometry_quality", default="standard", options=["standard", "detailed"], optional=True, advanced=True),
|
||||
],
|
||||
outputs=[
|
||||
@ -379,7 +372,6 @@ class TripoMultiviewToModelNode(IO.ComfyNode):
|
||||
"model_version",
|
||||
"texture",
|
||||
"pbr",
|
||||
"quad",
|
||||
"texture_quality",
|
||||
"geometry_quality",
|
||||
],
|
||||
@ -387,17 +379,16 @@ class TripoMultiviewToModelNode(IO.ComfyNode):
|
||||
expr="""
|
||||
(
|
||||
$isV14 := $contains(widgets.model_version,"v1.4");
|
||||
$isV3OrLater := $contains(widgets.model_version,"v3.");
|
||||
$withTexture := widgets.texture or widgets.pbr;
|
||||
$isHdTexture := (widgets.texture_quality = "detailed");
|
||||
$isDetailedGeometry := (widgets.geometry_quality = "detailed");
|
||||
$baseCredits :=
|
||||
$isV14 ? 30 : ($withTexture ? 30 : 20);
|
||||
$credits :=
|
||||
$baseCredits
|
||||
+ (widgets.quad ? 5 : 0)
|
||||
$credits := $isV14 ? 30 : (
|
||||
($withTexture ? 30 : 20)
|
||||
+ ($isHdTexture ? 10 : 0)
|
||||
+ ($isDetailedGeometry ? 20 : 0);
|
||||
{"type":"usd","usd": $round($credits * 0.01, 2)}
|
||||
+ (($isDetailedGeometry and $isV3OrLater) ? 20 : 0)
|
||||
);
|
||||
{"type":"usd","usd": $round($credits * 0.01, 2), "format": {"approximate": true}}
|
||||
)
|
||||
""",
|
||||
),
|
||||
@ -457,7 +448,7 @@ class TripoMultiviewToModelNode(IO.ComfyNode):
|
||||
geometry_quality=geometry_quality,
|
||||
texture_alignment=texture_alignment,
|
||||
face_limit=face_limit if face_limit != -1 else None,
|
||||
quad=quad,
|
||||
quad=None,
|
||||
),
|
||||
)
|
||||
return await poll_until_finished(cls, response, average_duration=80)
|
||||
@ -498,7 +489,7 @@ class TripoTextureNode(IO.ComfyNode):
|
||||
expr="""
|
||||
(
|
||||
$tq := widgets.texture_quality;
|
||||
{"type":"usd","usd": ($contains($tq,"detailed") ? 0.2 : 0.1)}
|
||||
{"type":"usd","usd": ($contains($tq,"detailed") ? 0.2 : 0.1), "format": {"approximate": true}}
|
||||
)
|
||||
""",
|
||||
),
|
||||
@ -555,7 +546,7 @@ class TripoRefineNode(IO.ComfyNode):
|
||||
is_api_node=True,
|
||||
is_output_node=True,
|
||||
price_badge=IO.PriceBadge(
|
||||
expr="""{"type":"usd","usd":0.3}""",
|
||||
expr="""{"type":"usd","usd":0.3, "format": {"approximate": true}}""",
|
||||
),
|
||||
)
|
||||
|
||||
@ -592,7 +583,7 @@ class TripoRigNode(IO.ComfyNode):
|
||||
is_api_node=True,
|
||||
is_output_node=True,
|
||||
price_badge=IO.PriceBadge(
|
||||
expr="""{"type":"usd","usd":0.25}""",
|
||||
expr="""{"type":"usd","usd":0.25, "format": {"approximate": true}}""",
|
||||
),
|
||||
)
|
||||
|
||||
@ -652,7 +643,7 @@ class TripoRetargetNode(IO.ComfyNode):
|
||||
is_api_node=True,
|
||||
is_output_node=True,
|
||||
price_badge=IO.PriceBadge(
|
||||
expr="""{"type":"usd","usd":0.1}""",
|
||||
expr="""{"type":"usd","usd":0.1, "format": {"approximate": true}}""",
|
||||
),
|
||||
)
|
||||
|
||||
@ -761,19 +752,10 @@ class TripoConversionNode(IO.ComfyNode):
|
||||
"face_limit",
|
||||
"texture_size",
|
||||
"texture_format",
|
||||
"force_symmetry",
|
||||
"flatten_bottom",
|
||||
"flatten_bottom_threshold",
|
||||
"pivot_to_center_bottom",
|
||||
"scale_factor",
|
||||
"with_animation",
|
||||
"pack_uv",
|
||||
"bake",
|
||||
"part_names",
|
||||
"fbx_preset",
|
||||
"export_vertex_colors",
|
||||
"export_orientation",
|
||||
"animate_in_place",
|
||||
],
|
||||
),
|
||||
expr="""
|
||||
@ -783,28 +765,16 @@ class TripoConversionNode(IO.ComfyNode):
|
||||
$flatThresh := (widgets.flatten_bottom_threshold != null) ? widgets.flatten_bottom_threshold : 0;
|
||||
$scale := (widgets.scale_factor != null) ? widgets.scale_factor : 1;
|
||||
$texFmt := (widgets.texture_format != "" ? widgets.texture_format : "jpeg");
|
||||
$part := widgets.part_names;
|
||||
$fbx := (widgets.fbx_preset != "" ? widgets.fbx_preset : "blender");
|
||||
$orient := (widgets.export_orientation != "" ? widgets.export_orientation : "default");
|
||||
$advanced :=
|
||||
widgets.quad or
|
||||
widgets.force_symmetry or
|
||||
widgets.flatten_bottom or
|
||||
widgets.pivot_to_center_bottom or
|
||||
widgets.with_animation or
|
||||
widgets.pack_uv or
|
||||
widgets.bake or
|
||||
widgets.export_vertex_colors or
|
||||
widgets.animate_in_place or
|
||||
($face != -1) or
|
||||
($texSize != 4096) or
|
||||
($flatThresh != 0) or
|
||||
($scale != 1) or
|
||||
($texFmt != "jpeg") or
|
||||
($part != "") or
|
||||
($fbx != "blender") or
|
||||
($orient != "default");
|
||||
{"type":"usd","usd": ($advanced ? 0.1 : 0.05)}
|
||||
($texFmt != "jpeg");
|
||||
{"type":"usd","usd": ($advanced ? 0.1 : 0.05), "format": {"approximate": true}}
|
||||
)
|
||||
""",
|
||||
),
|
||||
|
||||
@ -488,10 +488,30 @@ async def _diagnose_connectivity() -> dict[str, bool]:
|
||||
"api_accessible": False,
|
||||
}
|
||||
timeout = aiohttp.ClientTimeout(total=5.0)
|
||||
|
||||
# Probe Google and Baidu in parallel: Google is blocked by the GFW in mainland China, so a Baidu probe is required
|
||||
# to correctly detect that Chinese users with working internet do have working internet.
|
||||
internet_probe_urls = ("https://www.google.com", "https://www.baidu.com")
|
||||
|
||||
async with aiohttp.ClientSession(timeout=timeout) as session:
|
||||
with contextlib.suppress(ClientError, OSError):
|
||||
async with session.get("https://www.google.com") as resp:
|
||||
results["internet_accessible"] = resp.status < 500
|
||||
async def _probe(url: str) -> bool:
|
||||
try:
|
||||
async with session.get(url) as resp:
|
||||
return resp.status < 500
|
||||
except (ClientError, OSError, asyncio.TimeoutError):
|
||||
return False
|
||||
|
||||
probe_tasks = [asyncio.create_task(_probe(u)) for u in internet_probe_urls]
|
||||
try:
|
||||
for fut in asyncio.as_completed(probe_tasks):
|
||||
if await fut:
|
||||
results["internet_accessible"] = True
|
||||
break
|
||||
finally:
|
||||
for t in probe_tasks:
|
||||
if not t.done():
|
||||
t.cancel()
|
||||
await asyncio.gather(*probe_tasks, return_exceptions=True)
|
||||
if not results["internet_accessible"]:
|
||||
return results
|
||||
|
||||
|
||||
@ -11,6 +11,142 @@ from typing_extensions import override
|
||||
from comfy_api.latest import ComfyExtension, io
|
||||
import re
|
||||
|
||||
def video_latent_composite(destination, source, x, y, mask=None, multiplier=8, resize_source=False):
|
||||
# destination/source shape: [B, C, F, H, W]
|
||||
source = source.to(destination.device)
|
||||
|
||||
if resize_source:
|
||||
target_size = (source.shape[2], destination.shape[3], destination.shape[4])
|
||||
source = torch.nn.functional.interpolate(
|
||||
source,
|
||||
size=target_size,
|
||||
mode="trilinear",
|
||||
align_corners=False
|
||||
)
|
||||
|
||||
x_latent = x // multiplier
|
||||
y_latent = y // multiplier
|
||||
|
||||
if mask is None:
|
||||
mask = torch.ones_like(source)
|
||||
else:
|
||||
mask = mask.to(destination.device, copy=True)
|
||||
mask = mask.unsqueeze(0).unsqueeze(0)
|
||||
mask_target_size = (mask.shape[2], source.shape[3], source.shape[4])
|
||||
mask = torch.nn.functional.interpolate(
|
||||
mask,
|
||||
size=mask_target_size,
|
||||
mode="trilinear",
|
||||
align_corners=False
|
||||
)
|
||||
|
||||
dst_h, dst_w = destination.shape[3], destination.shape[4]
|
||||
src_h, src_w = source.shape[3], source.shape[4]
|
||||
|
||||
visible_h = max(0, min(y_latent + src_h, dst_h) - max(0, y_latent))
|
||||
visible_w = max(0, min(x_latent + src_w, dst_w) - max(0, x_latent))
|
||||
|
||||
if visible_h <= 0 or visible_w <= 0:
|
||||
return destination
|
||||
|
||||
src_top = max(0, -y_latent)
|
||||
src_left = max(0, -x_latent)
|
||||
dst_top = max(0, y_latent)
|
||||
dst_left = max(0, x_latent)
|
||||
|
||||
m = mask[:, :, :, src_top:src_top+visible_h, src_left:src_left+visible_w]
|
||||
s = source[:, :, :, src_top:src_top+visible_h, src_left:src_left+visible_w]
|
||||
d = destination[:, :, :, dst_top:dst_top+visible_h, dst_left:dst_left+visible_w]
|
||||
|
||||
destination[:, :, :, dst_top:dst_top+visible_h, dst_left:dst_left+visible_w] = (m * s) + ((1.0 - m) * d)
|
||||
|
||||
return destination
|
||||
|
||||
def time_to_move_sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, latent_mask, denoise=1.0, start_step=None, time_to_move_last_step=None, last_step=None, force_full_denoise=False, noise_mask=None, sigmas=None, callback=None, disable_pbar=False, seed=None):
|
||||
|
||||
sampler = comfy.samplers.KSampler(model, steps=steps, device=model.load_device, sampler=sampler_name, scheduler=scheduler, denoise=denoise, model_options=model.model_options)
|
||||
model_sampling = model.get_model_object("model_sampling")
|
||||
process_latent_out = model.get_model_object("process_latent_out")
|
||||
process_latent_in = model.get_model_object("process_latent_in")
|
||||
|
||||
reference_latent_image = latent_image.clone()
|
||||
|
||||
reference_sigmas = sampler.sigmas
|
||||
reference_noise = noise.clone()
|
||||
|
||||
if last_step == None or last_step > steps:
|
||||
last_step = steps
|
||||
|
||||
if time_to_move_last_step == None or time_to_move_last_step > last_step:
|
||||
time_to_move_last_step = last_step
|
||||
|
||||
if start_step == None:
|
||||
start_step = 0
|
||||
|
||||
total_iterations = min(last_step, steps) - start_step
|
||||
if total_iterations <= 0:
|
||||
return latent_image.to(
|
||||
device=comfy.model_management.intermediate_device(),
|
||||
dtype=comfy.model_management.intermediate_dtype(),
|
||||
)
|
||||
|
||||
for i in range(total_iterations):
|
||||
if i > 0:
|
||||
#don't add new noise to samples after first step taken
|
||||
noise = torch.zeros(latent_image.size(), dtype=latent_image.dtype, layout=latent_image.layout, device="cpu")
|
||||
|
||||
temp_start = start_step + i
|
||||
|
||||
if temp_start < last_step - 1:
|
||||
temp_force_full_denoise = False
|
||||
else:
|
||||
temp_force_full_denoise = force_full_denoise
|
||||
|
||||
samples = sampler.sample(noise, positive, negative, cfg=cfg, latent_image=latent_image, start_step=temp_start, last_step=temp_start + 1, force_full_denoise=temp_force_full_denoise, denoise_mask=noise_mask, sigmas=sigmas, callback=callback, disable_pbar=disable_pbar, seed=seed)
|
||||
|
||||
if temp_start < time_to_move_last_step:
|
||||
scale = reference_sigmas[temp_start + 1].to(noise.device)
|
||||
|
||||
if torch.count_nonzero(reference_latent_image) > 0: #Don't shift the empty latent image.
|
||||
noisy = model_sampling.noise_scaling(scale, reference_noise, process_latent_in(reference_latent_image))
|
||||
noisy = model_sampling.inverse_noise_scaling(scale, noisy)
|
||||
noisy = process_latent_out(noisy)
|
||||
else:
|
||||
noisy = reference_latent_image
|
||||
|
||||
noisy.to(samples.device)
|
||||
|
||||
samples = video_latent_composite(samples, noisy, 0, 0, latent_mask, multiplier=1, resize_source=True)
|
||||
|
||||
latent_image = samples
|
||||
|
||||
samples = samples.to(device=comfy.model_management.intermediate_device(), dtype=comfy.model_management.intermediate_dtype())
|
||||
return samples
|
||||
|
||||
|
||||
def time_to_move_common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent, latent_mask, denoise=1.0, disable_noise=False, start_step=None, time_to_move_last_step = None, last_step=None, force_full_denoise=False):
|
||||
latent_image = latent["samples"]
|
||||
latent_image = comfy.sample.fix_empty_latent_channels(model, latent_image, latent.get("downscale_ratio_spacial", None))
|
||||
|
||||
if disable_noise:
|
||||
noise = torch.zeros(latent_image.size(), dtype=latent_image.dtype, layout=latent_image.layout, device="cpu")
|
||||
else:
|
||||
batch_inds = latent["batch_index"] if "batch_index" in latent else None
|
||||
noise = comfy.sample.prepare_noise(latent_image, seed, batch_inds)
|
||||
|
||||
noise_mask = None
|
||||
if "noise_mask" in latent:
|
||||
noise_mask = latent["noise_mask"]
|
||||
|
||||
callback = latent_preview.prepare_callback(model, steps)
|
||||
disable_pbar = not comfy.utils.PROGRESS_BAR_ENABLED
|
||||
samples = time_to_move_sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, latent_mask,
|
||||
denoise=denoise, start_step=start_step, time_to_move_last_step = time_to_move_last_step, last_step=last_step,
|
||||
force_full_denoise=force_full_denoise, noise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=seed)
|
||||
out = latent.copy()
|
||||
out.pop("downscale_ratio_spacial", None)
|
||||
out["samples"] = samples
|
||||
return (out, )
|
||||
|
||||
class BasicScheduler(io.ComfyNode):
|
||||
@classmethod
|
||||
@ -978,6 +1114,46 @@ class SamplerCustomAdvanced(io.ComfyNode):
|
||||
return io.NodeOutput(out, out_denoised)
|
||||
|
||||
sample = execute
|
||||
|
||||
class TimeToMoveKSamplerAdvanced(io.ComfyNode):
|
||||
@classmethod
|
||||
|
||||
def define_schema(cls):
|
||||
return io.Schema(
|
||||
node_id="TimeToMoveKSamplerAdvanced",
|
||||
category="sampling/time_to_move",
|
||||
inputs=[
|
||||
io.Model.Input("model"),
|
||||
io.Combo.Input("add_noise", options=["enable", "disable"], advanced=True),
|
||||
io.Int.Input("noise_seed", default=0, min=0, max=0xffffffffffffffff, control_after_generate=True),
|
||||
io.Int.Input("steps", default=20, min=1, max=10000),
|
||||
io.Float.Input("cfg", default=8.0, min=0.0, max=100.0, step=0.1, round=0.01),
|
||||
io.Combo.Input("sampler_name", options = comfy.samplers.KSampler.SAMPLERS),
|
||||
io.Combo.Input("scheduler", options = comfy.samplers.KSampler.SCHEDULERS),
|
||||
io.Conditioning.Input("positive"),
|
||||
io.Conditioning.Input("negative"),
|
||||
io.Latent.Input("latent_image"),
|
||||
io.Mask.Input("latent_mask", tooltip = "Make sure mask is the same length as the latents rather than the original video."),
|
||||
io.Int.Input("start_at_step", default = 0, min = 0, max = 10000, advanced = True, tooltip = "Generally should set at a step greater than 0."),
|
||||
io.Int.Input("time_to_move_end_at_step", default = 0, min = 0, max = 10000, advanced = True, tooltip = "Generally should set at a step greater than 0 and less than total number of steps."),
|
||||
io.Int.Input("end_at_step", default = 10000, min = 0, max = 10000, advanced = True, tooltip = "Use just like typical end_at_step with normal KSamplerAdvanced"),
|
||||
io.Combo.Input("return_with_leftover_noise", options=["disable", "enable"], advanced = True),
|
||||
],
|
||||
outputs=[
|
||||
io.Latent.Output(display_name="latent"),
|
||||
]
|
||||
)
|
||||
|
||||
@classmethod
|
||||
def execute(cls, model, add_noise, noise_seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, latent_mask, start_at_step, time_to_move_end_at_step, end_at_step, return_with_leftover_noise, denoise=1.0) -> io.NodeOutput:
|
||||
force_full_denoise = True
|
||||
if return_with_leftover_noise == "enable":
|
||||
force_full_denoise = False
|
||||
disable_noise = False
|
||||
if add_noise == "disable":
|
||||
disable_noise = True
|
||||
|
||||
return time_to_move_common_ksampler(model, noise_seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, latent_mask, denoise=denoise, disable_noise=disable_noise, start_step=start_at_step, time_to_move_last_step = time_to_move_end_at_step, last_step=end_at_step, force_full_denoise=force_full_denoise)
|
||||
|
||||
class AddNoise(io.ComfyNode):
|
||||
@classmethod
|
||||
@ -1087,6 +1263,7 @@ class CustomSamplersExtension(ComfyExtension):
|
||||
DisableNoise,
|
||||
AddNoise,
|
||||
SamplerCustomAdvanced,
|
||||
TimeToMoveKSamplerAdvanced,
|
||||
ManualSigmas,
|
||||
]
|
||||
|
||||
|
||||
@ -57,6 +57,42 @@ def composite(destination, source, x, y, mask = None, multiplier = 8, resize_sou
|
||||
|
||||
return destination
|
||||
|
||||
def convert_rgb_mask_to_latent_mask(
|
||||
mask: torch.Tensor,
|
||||
k: int,
|
||||
spatial_downsample_h: int,
|
||||
spatial_downsample_w: int
|
||||
) -> torch.Tensor:
|
||||
"""
|
||||
Converts [T, H, W] mask to [T_latent, H_latent, W_latent].
|
||||
Handles non-square spatial downsampling.
|
||||
"""
|
||||
# 1. Temporal Sampling
|
||||
# Select first frame and every k-th frame thereafter
|
||||
mask0 = mask[0:1]
|
||||
mask1 = mask[1::k]
|
||||
sampled = torch.cat([mask0, mask1], dim=0) # [T_latent, H, W]
|
||||
|
||||
# 2. Prepare for Spatial Interpolation
|
||||
# Shape: [Batch=1, Channels=1, Depth=T_latent, Height=H, Width=W]
|
||||
sampled = sampled.unsqueeze(0).unsqueeze(0)
|
||||
|
||||
# 3. Calculate New Spatial Dimensions
|
||||
h_latent = sampled.shape[-2] // spatial_downsample_h
|
||||
w_latent = sampled.shape[-1] // spatial_downsample_w
|
||||
|
||||
# 4. Interpolate
|
||||
# We maintain the temporal count (sampled.shape[2])
|
||||
# but resize H and W independently
|
||||
pooled = torch.nn.functional.interpolate(
|
||||
sampled,
|
||||
size=(sampled.shape[2], h_latent, w_latent),
|
||||
mode="nearest"
|
||||
)
|
||||
|
||||
# 5. Return to [T_latent, H_latent, W_latent]
|
||||
return pooled.squeeze(0).squeeze(0)
|
||||
|
||||
class LatentCompositeMasked(IO.ComfyNode):
|
||||
@classmethod
|
||||
def define_schema(cls):
|
||||
@ -84,8 +120,7 @@ class LatentCompositeMasked(IO.ComfyNode):
|
||||
return IO.NodeOutput(output)
|
||||
|
||||
composite = execute # TODO: remove
|
||||
|
||||
|
||||
|
||||
class ImageCompositeMasked(IO.ComfyNode):
|
||||
@classmethod
|
||||
def define_schema(cls):
|
||||
@ -418,6 +453,30 @@ class ThresholdMask(IO.ComfyNode):
|
||||
|
||||
image_to_mask = execute # TODO: remove
|
||||
|
||||
class RGBMaskToLatentMask(IO.ComfyNode):
|
||||
@classmethod
|
||||
def define_schema(cls):
|
||||
return IO.Schema(
|
||||
node_id="RGBMasktoLatentMask",
|
||||
search_aliases=["rgb mask to latent mask", "rgb mask", "latent mask"],
|
||||
description="Converts an RGB mask to a latent-space mask for use with causal Video VAEs (e.g., Wan).",
|
||||
category="latent",
|
||||
inputs=[
|
||||
IO.Mask.Input("mask", optional=False),
|
||||
IO.Vae.Input("vae", optional=False),
|
||||
],
|
||||
outputs=[IO.Mask.Output()],
|
||||
)
|
||||
|
||||
@classmethod
|
||||
def execute(cls, mask, vae) -> IO.NodeOutput:
|
||||
# Ensure we work on a copy of the mask to remain non-destructive
|
||||
mask_copy = mask.clone()
|
||||
downscale_ratio = vae.downscale_ratio
|
||||
if not isinstance(downscale_ratio, tuple) or len(downscale_ratio) < 3:
|
||||
raise ValueError("RGBMaskToLatentMask requires a causal Video VAE (e.g., Wan). The provided VAE does not have a compatible downscale_ratio.")
|
||||
k = (mask.shape[0] - 1) // (downscale_ratio[0](mask.shape[0]) - 1) if (downscale_ratio[0](mask.shape[0]) - 1) > 1 else 1
|
||||
return IO.NodeOutput(convert_rgb_mask_to_latent_mask(mask_copy, k, spatial_downsample_h = downscale_ratio[1], spatial_downsample_w = downscale_ratio[2]))
|
||||
|
||||
# Mask Preview - original implement from
|
||||
# https://github.com/cubiq/ComfyUI_essentials/blob/9d9f4bedfc9f0321c19faf71855e228c93bd0dc9/mask.py#L81
|
||||
@ -459,6 +518,7 @@ class MaskExtension(ComfyExtension):
|
||||
FeatherMask,
|
||||
GrowMask,
|
||||
ThresholdMask,
|
||||
RGBMaskToLatentMask,
|
||||
MaskPreview,
|
||||
]
|
||||
|
||||
|
||||
4712
openapi.yaml
4712
openapi.yaml
File diff suppressed because it is too large
Load Diff
Loading…
Reference in New Issue
Block a user