526 Commits

Author	SHA1	Message	Date
Yousef Rafat	ee2b66a2f2	small fixes	2026-02-19 00:48:26 +02:00
Yousef Rafat	b5feac202c	.	2026-02-18 22:01:09 +02:00
Yousef R. Gamaleldin	9257acaae3	Merge branch 'master' into trellis2	2026-02-18 21:52:39 +02:00
chaObserv	44f8598521	Fix anima LLM adapter forward when manual cast (#12504)	2026-02-17 07:56:44 -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
Yousef Rafat	c14317d6e0	postprocessing node fixes + model small fixes	2026-02-17 00:10:48 +02:00
comfyanonymous	88e6370527	Remove workaround for old pytorch. (#12480)	2026-02-15 20:43:53 -05:00
Yousef Rafat	91fa563b21	rewriting conditioning logic + model code addition	2026-02-16 01:53:53 +02:00
krigeta	dc9822b7df	Add working Qwen 2512 ControlNet (Fun ControlNet) support (#12359)	2026-02-13 22:23:52 -05:00
comfyanonymous	726af73867	Fix some custom nodes. (#12455)	2026-02-13 20:21:10 -05:00
comfyanonymous	e1add563f9	Use torch RMSNorm for flux models and refactor hunyuan video code. (#12432)	2026-02-13 15:35:13 -05:00
Yousef Rafat	92aa058a58	.	2026-02-13 21:05:59 +02:00
Yousef Rafat	0e51bee64f	more reliable detection	2026-02-13 00:10:25 +02:00
Yousef Rafat	b7764479c2	debugging	2026-02-11 20:33:59 +02:00
comfyanonymous	76a7fa96db	Make built in lora training work on anima. (#12402)	2026-02-10 22:04:32 -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
Yousef Rafat	f4059c189e	dinov3 fixes + other	2026-02-11 01:27:54 +02:00
Yousef Rafat	704e1b5462	small bug fixes	2026-02-09 00:41:01 +02:00
comfyanonymous	039955c527	Some fixes to previous pr. (#12339)	2026-02-06 20:14:52 -05:00
tdrussell	6a26328842	Support fp16 for Cosmos-Predict2 and Anima (#12249)	2026-02-06 20:12:15 -05:00
comfyanonymous	eba6c940fd	Make ace step 1.5 base model work properly with default workflow. (#12337)	2026-02-06 19:14:56 -05:00
Yousef Rafat	955c00ee38	post-process node	2026-02-06 23:54:27 +02:00
Yousef Rafat	64a52f5585	checkpoint	2026-02-06 20:35:33 +02:00
Yousef Rafat	cdd7ced1e8	model fixes	2026-02-06 19:28:49 +02:00
comfyanonymous	458292fef0	Fix some lowvram stuff with ace step 1.5 (#12312)	2026-02-05 19:15:04 -05:00
comfyanonymous	6555dc65b8	Make ace step 1.5 work without the llm. (#12311)	2026-02-05 16:43:45 -05:00
Yousef Rafat	f2c0320fe8	fixes to vae and cumesh impl.	2026-02-05 17:19:57 +02:00
comfyanonymous	a50c32d63f	Disable sage attention on ace step 1.5 (#12297)	2026-02-04 22:15:30 -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
Yousef Rafat	cd0f7ba64e	apply rope and optimized attention	2026-02-05 02:34:08 +02:00
Yousef Rafat	3002708fe3	needed updates	2026-02-04 14:15:00 +02:00
Yousef Rafat	6624939505	structure model	2026-02-03 22:40:54 +02:00
Yousef Rafat	d6573fd26d	model init working	2026-02-03 21:10:20 +02:00
comfyanonymous	3c1a1a2df8	Basic support for the ace step 1.5 model. (#12237)	2026-02-03 00:06:18 -05:00
Yousef Rafat	f76e3a11b5	..	2026-02-02 21:27:15 +02:00
Yousef R. Gamaleldin	f1d25a460c	Merge branch 'master' into trellis2	2026-02-02 18:13:58 +02:00
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
comfyanonymous	b8f848bfe3	Fix model not working with any res. (#12186)	2026-01-31 00:12:48 -05:00
Yousef R. Gamaleldin	6ea2e5b288	init	2026-01-30 23:34:48 +02:00
rattus	6516ab335d	wan-vae: Switch off feature cache for single frame (#12090) ... The code throughout is None safe to just skip the feature cache saving step if none. Set it none in single frame use so qwen doesn't burn VRAM on the unused cache.	2026-01-26 19:40:19 -05:00
comfyanonymous	635406e283	Only enable fp16 on z image models that actually support it. (#12065)	2026-01-24 22:32:28 -05:00
rattus	4e6a1b66a9	speed up and reduce VRAM of QWEN VAE and WAN (less so) (#12036) ... * ops: introduce autopad for conv3d This works around pytorch missing ability to causal pad as part of the kernel and avoids massive weight duplications for padding. * wan-vae: rework causal padding This currently uses F.pad which takes a full deep copy and is liable to be the VRAM peak. Instead, kick spatial padding back to the op and consolidate the temporal padding with the cat for the cache. * wan-vae: implement zero pad fast path The WAN VAE is also QWEN where it is used single-image. These convolutions are however zero padded 3d convolutions, which means the VAE is actually just 2D down the last element of the conv weight in the temporal dimension. Fast path this, to avoid adding zeros that then just evaporate in convoluton math but cost computation.	2026-01-23 19:56:14 -05:00
Jukka Seppänen	55bd606e92	LTX2: Refactor forward function for better VRAM efficiency and fix spatial inpainting (#12046) ... * Disable timestep embed compression when inpainting Spatial inpainting not compatible with the compression * Reduce crossattn peak VRAM * LTX2: Refactor forward function for better VRAM efficiency	2026-01-23 15:26:38 -05:00
Omri Marom	d7f3241bf6	qwen_image: propagate attention mask. (#11966)	2026-01-22 20:02:31 -05:00
rattus	0fd1b78736	Reduce LTX2 VAE VRAM consumption (#12028) ... * causal_video_ae: Remove attention ResNet This attention_head_dim argument does not exist on this constructor so this is dead code. Remove as generic attention mid VAE conflicts with temporal roll. * ltx-vae: consoldate causal/non-causal code paths * ltx-vae: add cache rolling adder * ltx-vae: use cached adder for resnet * ltx-vae: Implement rolling VAE Implement a temporal rolling VAE for the LTX2 VAE. Usually when doing temporal rolling VAEs you can just chunk on time relying on causality and cache behind you as you go. The LTX VAE is however non-causal. So go whole hog and implement per layer run ahead and backpressure between the decoder layers using recursive state beween the layers. Operations are ammended with temporal_cache_state{} which they can use to hold any state then need for partial execution. Convolutions cache their inputs behind the up to N-1 frames, and skip connections need to cache the mismatch between convolution input and output that happens due to missing future (non-causal) input. Each call to run_up() processes a layer accross a range on input that may or may not be complete. It goes depth first to process as much as possible to try and digest frames to the final output ASAP. If layers run out of input due to convolution losses, they simply return without action effectively applying back-pressure to the earlier layers. As the earlier layers do more work and caller deeper, the partial states are reconciled and output continues to digest depth first as much as possible. Chunking is done using a size quota rather than a fixed frame length and any layer can initiate chunking, and multiple layers can chunk at different granulatiries. This remove the old limitation of always having to process 1 latent frame to entirety and having to hold 8 full decoded frames as the VRAM peak.	2026-01-22 16:54:18 -05:00
Jukka Seppänen	16b9aabd52	Support Multi/InfiniteTalk (#10179) ... * re-init * Update model_multitalk.py * whitespace... * Update model_multitalk.py * remove print * this is redundant * remove import * Restore preview functionality * Move block_idx to transformer_options * Remove LoopingSamplerCustomAdvanced * Remove looping functionality, keep extension functionality * Update model_multitalk.py * Handle ref_attn_mask with separate patch to avoid having to always return q and k from self_attn * Chunk attention map calculation for multiple speakers to reduce peak VRAM usage * Update model_multitalk.py * Add ModelPatch type back * Fix for latest upstream * Use DynamicCombo for cleaner node Basically just so that single_speaker mode hides mask inputs and 2nd audio input * Update nodes_wan.py	2026-01-21 23:09:48 -05:00
comfyanonymous	abe2ec26a6	Support the Anima model. (#12012)	2026-01-21 19:44:28 -05:00
Ivan Zorin	965d0ed509	fix: remove normalization of audio in LTX Mel spectrogram creation (#11990) ... For LTX Audio VAE, remove normalization of audio during MEL spectrogram creation. This aligs inference with training and prevents loud audio from being attenuated.	2026-01-20 18:44:28 -05:00
comfyanonymous	8ccc0c94fa	Make omni stuff work on regular z image for easier testing. (#11985)	2026-01-20 00:32:00 -05:00
comfyanonymous	2108167f9f	Support zimage omni base model. (#11979)	2026-01-19 23:17:38 -05:00