50 Commits

Author	SHA1	Message	Date
Jukka Seppänen	8d6e53b556	Allow compiling VAE as well	2024-11-23 17:08:57 +02:00
kijai	895d3b83a4	Update model_loading.py	2024-11-21 03:05:51 +02:00
kijai	276a045a57	use selected load device as LoRA load device too	2024-11-21 02:46:07 +02:00
kijai	e52dc36bc5	Update model_loading.py	2024-11-20 21:28:53 +02:00
kijai	e5fc7c1bf3	Allow mixing Fun and not fun loras	2024-11-20 21:24:51 +02:00
kijai	e187cfe22f	Allow loading the "Rewards" LoRAs into 1.5 as well (for what it's worth)	2024-11-20 19:18:40 +02:00
kijai	b74aa75026	Don't use autocast with fp/bf16	2024-11-20 14:22:10 +02:00
kijai	ecd067260c	Add CogVideoX-Fun-V1.1-5b-Control ... https://huggingface.co/alibaba-pai/CogVideoX-Fun-V1.1-5b-Control	2024-11-20 01:23:54 +02:00
kijai	41a0f33381	Update model_loading.py	2024-11-19 20:27:31 +02:00
kijai	1cfe0835f5	fix GGUF loader	2024-11-19 20:23:47 +02:00
kijai	882faa6dea	add comfyui attention mode	2024-11-19 19:55:51 +02:00
kijai	516655b689	Update model_loading.py	2024-11-19 19:17:42 +02:00
kijai	67f2f6abb1	Merge branch 'refactor'	2024-11-19 19:16:39 +02:00
kijai	feeff366b5	update	2024-11-19 19:06:15 +02:00
kijai	6302e4b668	Allow orbit LoRAs with Fun models as well	2024-11-19 15:49:43 +02:00
kijai	a7646c0d6f	refactor ... - unify all pipelines into one - unify transformer model into one - separate VAE - add single file model loading	2024-11-19 03:04:22 +02:00
kijai	15aa68c95d	update	2024-11-17 00:48:01 +02:00
kijai	4374273138	possible compile fixes	2024-11-16 22:18:12 +02:00
kijai	bececf0189	some experimental optimizations	2024-11-16 17:32:31 +02:00
kijai	75e98906a3	rotary embed fix ... 25a9e1c567	2024-11-15 02:37:44 +02:00
kijai	0bd3da569e	code cleanup ... codebase getting too bloated: drop PAB support in favor of FasterCache drop temporal tilling in favor of FreeNoise	2024-11-14 19:54:52 +02:00
kijai	e8a289112f	fix VAE scaling (again)	2024-11-13 15:37:45 +02:00
kijai	ba2dbfbeb4	fixes	2024-11-12 11:29:50 +02:00
kijai	db697fea11	Update model_loading.py	2024-11-12 00:22:26 +02:00
kijai	00fde5ebce	allow fused loras with fp8	2024-11-11 23:36:48 +02:00
kijai	ea0273c8ec	VAE fix, allow using fp32 VAE	2024-11-11 22:34:43 +02:00
kijai	6d4c99e77d	Update model_loading.py	2024-11-11 19:58:17 +02:00
kijai	ca63f5dade	update	2024-11-11 01:19:11 +02:00
kijai	87ed4a56cf	Update model_loading.py	2024-11-10 18:19:35 +02:00
kijai	ea5ee0b017	GGUF Q4 works	2024-11-10 18:13:44 +02:00
kijai	fb246f95ef	attention compile works with higher cache_size_limit	2024-11-09 22:56:50 +02:00
kijai	a630bb3314	update	2024-11-09 22:17:10 +02:00
Jukka Seppänen	909d7026f3	Update model_loading.py	2024-11-09 20:20:15 +02:00
kijai	7162d1040d	Update model_loading.py	2024-11-09 18:14:48 +02:00
kijai	75aa19b4e1	Update model_loading.py	2024-11-09 18:12:03 +02:00
kijai	634c22db50	sageattn	2024-11-09 17:05:55 +02:00
kijai	1c3aff9000	some cleanup	2024-11-09 15:15:10 +02:00
kijai	643bbc18c1	i2v	2024-11-09 12:13:52 +02:00
Jukka Seppänen	2eb9b81d27	fp8	2024-11-09 04:41:07 +02:00
Jukka Seppänen	9a64e1ae5e	Update model_loading.py	2024-11-09 04:16:46 +02:00
Jukka Seppänen	b563994afc	finally ... works	2024-11-09 04:02:36 +02:00
kijai	806a0fa1d6	Update model_loading.py	2024-11-08 21:31:31 +02:00
kijai	2074ba578e	doesn't work yet	2024-11-08 21:24:20 +02:00
kijai	f7a88cbd56	Update model_loading.py	2024-11-08 21:23:29 +02:00
kijai	4c2ce52f57	Update model_loading.py	2024-11-08 17:30:51 +02:00
kijai	1cc6e1f070	use diffusers LoRA loading to support fusing for DimensionX LoRAs ... https://github.com/wenqsun/DimensionX	2024-11-08 14:24:32 +02:00
kijai	07defb52b6	Add compile_args node	2024-11-07 23:17:37 +02:00
kijai	9cce8015d3	Update model_loading.py	2024-11-07 23:02:52 +02:00
kijai	57262d9762	Update model_loading.py	2024-10-31 01:04:43 +02:00
kijai	5ba9b1d634	code cleanup	2024-10-30 21:30:03 +02:00