716 Commits

Author	SHA1	Message	Date
rattus	5f41584e96	Disable dynamic_vram when weight hooks applied (#12653) ... * sd: add support for clip model reconstruction * nodes: SetClipHooks: Demote the dynamic model patcher * mp: Make dynamic_disable more robust The backup need to not be cloned. In addition add a delegate object to ModelPatcherDynamic so that non-cloning code can do ModelPatcherDynamic demotion * sampler_helpers: Demote to non-dynamic model patcher when hooking * code rabbit review comments	2026-02-28 16:50:18 -05:00
Jukka Seppänen	1f6744162f	feat: Support SCAIL WanVideo model (#12614)	2026-02-28 16:49:12 -05:00
Christian Byrne	80d49441e5	refactor: use AspectRatio enum members as ASPECT_RATIOS dict keys (#12689) ... Amp-Thread-ID: https://ampcode.com/threads/T-019ca1cb-0150-7549-8b1b-6713060d3408 Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>	2026-02-27 20:53:46 -08:00
comfyanonymous	94f1a1cc9d	Limit overlap in image tile and combine nodes to prevent issues. (#12688)	2026-02-27 20:16:24 -05:00
Christian Byrne	1f1ec377ce	feat: add ResolutionSelector node for aspect ratio and megapixel-based resolution calculation (#12199) ... Amp-Thread-ID: https://ampcode.com/threads/T-019c179e-cd8c-768f-ae66-207c7a53c01d Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>	2026-02-27 09:13:57 -08:00
pythongosssss	0a7f8e11b6	fix torch.cat requiring inputs to all be same dimensions (#12673)	2026-02-27 08:13:24 -08:00
Jukka Seppänen	c7f7d52b68	feat: Support SDPose-OOD (#12661)	2026-02-26 19:59:05 -05:00
pythongosssss	38ca94599f	pyopengl-accelerate can cause object to be numpy ints instead of bare ints, which the glDeleteTextures function does not accept, explicitly cast to int (#12650)	2026-02-26 03:07:35 -08:00
Christian Byrne	74b5a337dc	fix: move essentials_category to correct replacement nodes (#12568) ... Move essentials_category from deprecated/incorrect nodes to their replacements: - ImageBatch → BatchImagesNode (ImageBatch is deprecated) - Blur → removed (should use subgraph blueprint) - GetVideoComponents → Video Slice Amp-Thread-ID: https://ampcode.com/threads/T-019c8340-4da2-723b-a09f-83895c5bbda5	2026-02-26 01:00:32 -08:00
Tavi Halperin	a4522017c5	feat: per-guide attention strength control in self-attention (#12518) ... Implements per-guide attention attenuation via log-space additive bias in self-attention. Each guide reference tracks its own strength and optional spatial mask in conditioning metadata (guide_attention_entries).	2026-02-26 01:25:23 -05:00
Jukka Seppänen	907e5dcbbf	initial FlowRVS support (#12637)	2026-02-25 23:38:46 -05:00
rattus	3ebe1ac22e	Disable dynamic_vram when using torch compiler (#12612) ... * mp: attach re-construction arguments to model patcher When making a model-patcher from a unet or ckpt, attach a callable function that can be called to replay the model construction. This can be used to deep clone model patcher WRT the actual model. Originally written by Kosinkadink f4b99bc623 * mp: Add disable_dynamic clone argument Add a clone argument that lets a caller clone a ModelPatcher but disable dynamic to demote the clone to regular MP. This is useful for legacy features where dynamic_vram support is missing or TBD. * torch_compile: disable dynamic_vram This is a bigger feature. Disable for the interim to preserve functionality.	2026-02-24 19:13:46 -05:00
comfyanonymous	9b1c63eb69	Add SplitImageToTileList and ImageMergeTileList nodes. (#12599) ... With these you can split an image into tiles, do operations and then combine it back to a single image.	2026-02-23 21:01:17 -05:00
Christian Byrne	ee72752162	Add category to Normalized Attention Guidance node (#12565)	2026-02-21 19:51:21 -05:00
pythongosssss	602b2505a4	add support for pyopengl < 3.1.4 where the size parameter does not exist (#12555)	2026-02-21 06:14:57 -08:00
Christian Byrne	04a55d5019	fix: swap essentials_category from CLIPTextEncode to PrimitiveStringMultiline (#12553) ... Remove CLIPTextEncode from Basics essentials category and add PrimitiveStringMultiline (String Multiline) in its place. Amp-Thread-ID: https://ampcode.com/threads/T-019c7efb-d916-7244-8c43-77b615ba0622 Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>	2026-02-20 23:46:46 -08:00
Christian Byrne	5fb8f06495	feat: add essential subgraph blueprints (#12552) ... Add 24 non-cloud essential blueprints from comfyui-wiki/Subgraph-Blueprints. These cover common workflows: text/image/video generation, editing, inpainting, outpainting, upscaling, depth maps, pose, captioning, and more. Cloud-only blueprints (5) are excluded and will be added once client-side distribution filtering lands. Amp-Thread-ID: https://ampcode.com/threads/T-019c6f43-6212-7308-bea6-bfc35a486cbf	2026-02-20 23:40:13 -08:00
pythongosssss	96d6bd1a4a	Add GLSL shader node using PyOpenGL (#12148) ... * adds support for executing simple glsl shaders using moderngl package * tidy * Support multiple outputs * Try fix build * fix casing * fix line endings * convert to using PyOpenGL and glfw * remove cpu support * tidy * add additional support for egl & osmesa backends * fix ci perf: only read required outputs * add diagnostics, update mac initialization * GLSL glueprints + node fixes (#12492) * Add image operation blueprints * Add channels * Add glow * brightness/contrast * hsb * add glsl shader update system * shader nit iteration * add multipass for faster blur * more fixes * rebuild blueprints * print -> logger * Add edge preserving blur * fix: move _initialized flag to end of GLContext.__init__ Prevents '_vao' attribute error when init fails partway through and subsequent calls skip initialization due to early _initialized flag. * update valid ranges - threshold 0-100 - step 0+ * fix value ranges * rebuild node to remove extra inputs * Fix gamma step * clamp saturation in colorize instead of wrapping * Fix crash on 1x1 px images * rework description * remove unnecessary f Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com> Co-authored-by: Hunter Senft-Grupp <hunter@comfy.org>	2026-02-19 23:22:13 -05:00
comfyanonymous	5f2117528a	Force min length 1 when tokenizing for text generation. (#12538)	2026-02-19 22:57:44 -05:00
Christian Byrne	4d172e9ad7	feat: mark 429 widgets as advanced for collapsible UI (#12197) ... * feat: mark 429 widgets as advanced for collapsible UI Mark widgets as advanced across core, comfy_extras, and comfy_api_nodes to support the new collapsible advanced inputs section in the frontend. Changes: - 267 advanced markers in comfy_extras/ - 162 advanced markers in comfy_api_nodes/ - All files pass python3 -m py_compile verification Widgets marked advanced (hidden by default): - Scheduler internals: sigma_max, sigma_min, rho, mu, beta, alpha - Sampler internals: eta, s_noise, order, rtol, atol, h_init, pcoeff, etc. - Memory optimization: tile_size, overlap, temporal_size, temporal_overlap - Pipeline controls: add_noise, start_at_step, end_at_step - Timing controls: start_percent, end_percent - Layer selection: stop_at_clip_layer, layers, block_number - Video encoding: codec, crf, format - Device/dtype: device, noise_device, dtype, weight_dtype Widgets kept basic (always visible): - Core params: strength, steps, cfg, denoise, seed, width, height - Model selectors: ckpt_name, lora_name, vae_name, sampler_name - Common controls: upscale_method, crop, batch_size, fps, opacity Related: frontend PR #11939 Amp-Thread-ID: https://ampcode.com/threads/T-019c1734-6b61-702e-b333-f02c399963fc * fix: remove advanced=True from DynamicCombo.Input (unsupported) Amp-Thread-ID: https://ampcode.com/threads/T-019c1734-6b61-702e-b333-f02c399963fc * fix: address review - un-mark model merge, video, image, and training node widgets as advanced Per comfyanonymous review: - Model merge arguments should not be advanced (all 14 model-specific merge classes) - SaveAnimatedWEBP lossless/quality/method should not be advanced - SaveWEBM/SaveVideo codec/crf/format should not be advanced - TrainLoraNode options should not be advanced (7 inputs) Amp-Thread-ID: https://ampcode.com/threads/T-019c322b-a3a8-71b7-9962-d44573ca6352 * fix: un-mark batch_size and webcam width/height as advanced (should stay basic) Amp-Thread-ID: https://ampcode.com/threads/T-019c3236-1417-74aa-82a3-bcb365fbe9d1 --------- Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>	2026-02-19 19:20:02 -08:00
Yourz	5632b2df9d	feat: add essentials_category (#12357) ... * feat: add essentials_category field to node schema Amp-Thread-ID: https://ampcode.com/threads/T-019c2b25-cd90-7218-9071-03cb46b351b3 * feat: add ESSENTIALS_CATEGORY to core nodes Marked nodes: - Basic: LoadImage, SaveImage, LoadVideo, SaveVideo, Load3D, CLIPTextEncode - Image Tools: ImageScale, ImageInvert, ImageBatch, ImageCrop, ImageRotate, ImageBlur - Image Tools/Preprocessing: Canny - Image Generation: LoraLoader - Audio: LoadAudio, SaveAudio Amp-Thread-ID: https://ampcode.com/threads/T-019c2b25-cd90-7218-9071-03cb46b351b3 * Add ESSENTIALS_CATEGORY to more nodes - SaveGLB (Basic) - GetVideoComponents (Video Tools) - TencentTextToModelNode, TencentImageToModelNode (3D) - RecraftRemoveBackgroundNode (Image Tools) - KlingLipSyncAudioToVideoNode (Video Generation) - OpenAIChatNode (Text Generation) - StabilityTextToAudio (Audio) Amp-Thread-ID: https://ampcode.com/threads/T-019c2b69-81c1-71c3-8096-450a39e20910 * fix: correct essentials category for Canny node Amp-Thread-ID: https://ampcode.com/threads/T-019c7303-ab53-7341-be76-a5da1f7a657e Co-authored-by: Amp <amp@ampcode.com> * refactor: replace essentials_category string literals with constants Amp-Thread-ID: https://ampcode.com/threads/T-019c7303-ab53-7341-be76-a5da1f7a657e Co-authored-by: Amp <amp@ampcode.com> * refactor: revert constants, use string literals for essentials_category Amp-Thread-ID: https://ampcode.com/threads/T-019c7303-ab53-7341-be76-a5da1f7a657e Co-authored-by: Amp <amp@ampcode.com> * fix: update basics --------- Co-authored-by: bymyself <cbyrne@comfy.org> Co-authored-by: Amp <amp@ampcode.com> Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>	2026-02-19 19:00:26 -08:00
Jukka Seppänen	6d11cc7354	feat: Add basic text generation support with native models, initially supporting Gemma3 (#12392)	2026-02-18 20:49:43 -05:00
comfyanonymous	f262444dd4	Add simple 3 band equalizer node for audio. (#12519)	2026-02-18 18:36:35 -05:00
Terry Jia	8ad38d2073	BBox widget (#11594) ... * Boundingbox widget * code improve --------- Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com> Co-authored-by: Christian Byrne <cbyrne@comfy.org>	2026-02-17 17:13:39 -08:00
comfyanonymous	18927538a1	Implement NAG on all the models based on the Flux code. (#12500) ... Use the Normalized Attention Guidance node. Flux, Flux2, Klein, Chroma, Chroma radiance, Hunyuan Video, etc..	2026-02-16 23:30:34 -05:00
rattus	ecd2a19661	Fix lora Extraction in offload conditions (+ dynamic_vram mode) (#12479) ... * lora_extract: Add a trange If you bite off more than your GPU can chew, this kinda just hangs. Give a rough indication of progress counting the weights in a trange. * lora_extract: Support on-the-fly patching Use the on-the-fly approach from the regular model saving logic for lora extraction too. Switch off force_cast_weights accordingly. This gets extraction working in dynamic vram while also supporting extraction on GPU offloaded.	2026-02-15 20:28:51 -05:00
Jedrzej Kosinski	596ed68691	Node Replacement API (#12014)	2026-02-15 02:12:30 -08:00
rattus	8902907d7a	dynamic_vram: Training fixes (#12442)	2026-02-13 15:29:37 -05:00
askmyteapot	e5ae670a40	Update ace15.py to allow min_p sampling (#12373)	2026-02-11 20:28:48 -05:00
Kohaku-Blueleaf	cdcf4119b3	[Trainer] training with proper offloading (#12189) ... * Fix bypass dtype/device moving * Force offloading mode for training * training context var * offloading implementation in training node * fix wrong input type * Support bypass load lora model, correct adapter/offloading handling	2026-02-10 21:45:19 -05:00
AustinMroz	dbe70b6821	Add a VideoSlice node (#12107) ... * Base TrimVideo implementation * Raise error if as_trimmed call fails * Bigger max start_time, tooltips, and formatting * Count packets unless codec has subframes * Remove incorrect nested decode * Add null check for audio streams * Support non-strict duration * Added strict_duration bool to node definition * Empty commit for approval * Fix duration * Support 5.1 audio layout on save --------- Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>	2026-02-10 14:42:21 -08:00
comfyanonymous	a0302cc6a8	Make tonemap latent work on any dim latents. (#12363)	2026-02-08 21:16:40 -05:00
chaObserv	9bf5aa54db	Add search_aliases to sa-solver and seeds-2 node (#12327)	2026-02-07 17:38:51 -05:00
Jukka Seppänen	5ff4fdedba	Fix LazyCache (#12344)	2026-02-07 11:25:30 -08:00
Jukka Seppänen	a1c101f861	EasyCache: Support LTX2 (#12231)	2026-02-06 00:43:09 -05:00
AustinMroz	2b70ab9ad0	Add a Create List node (#12173)	2026-02-05 01:18:21 -05:00
comfyanonymous	35183543e0	Add VAE tiled decode node for audio. (#12299)	2026-02-05 01:12:04 -05:00
comfyanonymous	6125b80979	Add llm sampling options and make reference audio work on ace step 1.5 (#12295)	2026-02-04 21:29:22 -05:00
Alexander Piskun	e77b34dfea	add File3DAny output to Load3D node; extend SaveGLB to accept File3DAny as input (#12276) ... * add File3DAny output to Load3D node; extend SaveGLB node to accept File3DAny as input * fix(grammar): capitalize letter	2026-02-04 11:35:38 -08:00
Alexander Piskun	fb23935c11	feat(comfy_api): add basic 3D Model file types (#12129) ... * feat(comfy_api): add basic 3D Model file types * update Tripo nodes to use File3DGLB * update Rodin3D nodes to use File3DGLB * address PR review feedback: - Rename File3D parameter 'path' to 'source' - Convert File3D.data property to get_data() - Make .glb extension check case-insensitive in nodes_rodin.py - Restrict SaveGLB node to only accept File3DGLB * Fixed a bug in the Meshy Rig and Animation nodes * Fix backward compatability	2026-02-03 10:31:46 -08:00
comfyanonymous	3c1a1a2df8	Basic support for the ace step 1.5 model. (#12237)	2026-02-03 00:06:18 -05:00
rattus	f8acd9c402	Reduce RAM usage, fix VRAM OOMs, and fix Windows shared memory spilling with adaptive model loading (#11845)	2026-02-01 01:01:11 -05:00
pythongosssss	8aabe2403e	Add color type and Color to RGB Int node (#12145) ... * add color type and color to rgb int node * review fix for allowing output --------- Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>	2026-01-30 15:01:33 -08:00
comfyanonymous	bbe2c13a70	Make empty hunyuan latent 1.0 work with the 1.5 model. (#12171)	2026-01-29 23:52:22 -05:00
Jukka Seppänen	ad53e78f11	Fix Noise_EmptyNoise when using nested latents (#12089)	2026-01-26 19:25:00 -05:00
Tavi Halperin	bfe31d0b9d	IC-LoRA: support small grid (#12074)	2026-01-26 15:33:19 -05:00
comfyanonymous	26c5bbb875	Move nodes from previous PR into their own file. (#12066)	2026-01-24 23:02:32 -05:00
Kohaku-Blueleaf	a97c98068f	[Weight-adapter/Trainer] Bypass forward mode in Weight adapter system (#11958) ... * Add API of bypass forward module * bypass implementation * add bypass fwd into nodes list/trainer	2026-01-24 22:56:22 -05:00
pythongosssss	ed6002cb60	add support for kwargs inputs to allow arbitrary inputs from frontend (#12063) ... used to output selected combo index Co-authored-by: Jedrzej Kosinski <kosinkadink1@gmail.com>	2026-01-24 17:30:40 -08:00
comfyanonymous	9cf299a9f9	Make regular empty latent node work properly on flux 2 variants. (#12050)	2026-01-23 19:50:48 -05:00