64 Commits

Author	SHA1	Message	Date
kijai	ff49e1b01f	Add sageattn++ as selectable mode for easier testing	2025-07-01 12:26:27 +03:00
kijai	7afb0f906a	Update model_optimization_nodes.py	2025-06-16 00:35:53 +03:00
kijai	d584c711a3	Experimental NAG node	2025-06-15 22:05:24 +03:00
kijai	dfe7a909a4	Update model_optimization_nodes.py	2025-06-09 04:30:00 +03:00
kijai	e96a028254	Support VACE with TeaCache	2025-06-09 02:58:41 +03:00
kijai	87b8f85103	Fix Wan enhance a video disabling sageattention patch	2025-06-06 15:46:33 +03:00
kijai	07b804cb3f	Add TorchCompileModelWanVideoV2	2025-05-21 17:38:22 +03:00
kijai	16f60e53e5	Add TorchCompileModelFluxAdvancedV2 ... Utilizing new ComfyUI core functions that also support LoRAs without additional patches	2025-05-21 17:17:41 +03:00
kijai	bec42252c6	smarter sage patch	2025-05-07 19:08:25 +03:00
kijai	ca07b9dadc	Add ModelPatchTorchSettings	2025-05-07 19:00:45 +03:00
kijai	6a35eefdd5	Update model_optimization_nodes.py	2025-05-07 17:52:47 +03:00
kijai	bfb6d973fe	update CheckpointLoaderKJ	2025-05-07 17:30:38 +03:00
kijai	086ac449a2	Update model_optimization_nodes.py	2025-04-23 20:09:09 +03:00
kijai	8dac94d9d9	Update model_optimization_nodes.py	2025-04-23 20:08:10 +03:00
kijai	58668a44cc	Update model_optimization_nodes.py	2025-04-19 16:55:05 +03:00
kijai	9026379046	Use pre_run callback for applying sageattention patch ... It's still global patch, but now applied before using the model thus better allowing disabling it for unsupported models	2025-04-19 16:26:59 +03:00
kijai	0addfc6101	Maybe fix shape mismatch with SLG and I2V when doing batched cfg	2025-03-29 12:51:49 +02:00
kijai	916461c432	fix zero_init	2025-03-29 11:28:08 +02:00
kijai	58a06698f0	small tweaks	2025-03-28 17:24:44 +02:00
kijai	52c2e31a90	rename CFGZeroStar to avoid conflict	2025-03-26 12:29:56 +02:00
kijai	f77c0dc5d8	Update model_optimization_nodes.py	2025-03-26 01:14:16 +02:00
kijai	b1ec996ba3	Add CFGZeroStar ... https://github.com/WeichenFan/CFG-Zero-star/	2025-03-26 01:05:20 +02:00
kijai	a5bd3c86c8	Make TeaCache node error out instead of silently fail if ComfyUI isn't new enough to pass transformer_options	2025-03-19 10:13:08 +02:00
ozbayb	71351c0984	Fix for substep scheduling of patches with multiples conds	2025-03-17 13:51:06 -06:00
kijai	393ec896f7	Better compile compatibility with various patches ... Shouldn't drop compile when changing slg or enhance-a-video settings anymore	2025-03-17 14:11:30 +02:00
kijai	bb154eb71f	Fix TeaCache start/end	2025-03-17 09:48:06 +02:00
kijai	7c488a16ef	Add SkipLayerGuidanceWanVideo	2025-03-16 17:24:23 +02:00
kijai	6b7eeebe44	Update model_optimization_nodes.py	2025-03-16 13:15:47 +02:00
kijai	c19ad34916	update to match comfyui latest update	2025-03-14 18:42:31 +02:00
kijai	a5b39a8f1a	Update model_optimization_nodes.py	2025-03-13 11:10:05 +02:00
kijai	a4b9fd36da	Possible workaround for model patch order patch memory issue	2025-03-12 00:05:42 +02:00
kijai	51b9efe0a1	Update model_optimization_nodes.py	2025-03-11 19:46:10 +02:00
kijai	263961539e	Add WanVideoEnhanceAVideoKJ	2025-03-09 16:53:51 +02:00
kijai	665f59fae3	Include WanVideoWrapper for help popups	2025-03-08 12:15:32 +02:00
kijai	28d1fbda34	update TeaCache docs	2025-03-08 12:09:02 +02:00
kijai	79d9aab5e7	Update model_optimization_nodes.py	2025-03-08 10:49:27 +02:00
kijai	773fab5a07	Update model_optimization_nodes.py	2025-03-08 10:49:17 +02:00
kijai	68db110554	Exclude TeaCache from compile to avoid possible compile errors, make compiling whole model default for WanVideo	2025-03-07 16:35:30 +02:00
kijai	8ba503c0a5	Update model_optimization_nodes.py	2025-03-07 01:38:40 +02:00
kijai	fa6d20eeb3	add WanVideo TeaCache coefficients	2025-03-05 14:03:05 +02:00
kijai	11a3744c3d	Update model_optimization_nodes.py	2025-03-03 19:54:33 +02:00
kijai	60abdef03e	rename WanVideoTeaCache -> WanVideoTeaCacheKJ to avoid clash with wrapper	2025-03-03 16:55:56 +02:00
kijai	3b0d20532f	Update model_optimization_nodes.py	2025-03-03 15:40:17 +02:00
kijai	d00082f648	Add WanVideoTeaCache	2025-03-03 15:23:53 +02:00
kijai	9a15e22f5e	Update model_optimization_nodes.py	2025-03-01 12:41:50 +02:00
Jukka Seppänen	6a16be7f21	Merge pull request #202 from DocShotgun/main ... Fix nightly torch check for fp16 accumulation	2025-02-28 22:20:19 +02:00
kijai	46b76b8d7c	Update model_optimization_nodes.py	2025-02-28 20:59:57 +02:00
DocShotgun	9dc737e560	Fix nightly torch check for fp16 accumulation	2025-02-28 10:59:29 -08:00
kijai	ced72635c7	Add new options to the model loader	2025-02-28 19:27:35 +02:00
kijai	1a4259f052	TorchCompileModelWanVideo	2025-02-26 11:10:17 +02:00