kijai
d5399e8667
Merge branch 'main' into dev
2025-08-19 16:56:00 +03:00
kijai
345c286c86
Update utils.py
2025-08-19 16:55:51 +03:00
kijai
10e1dd68ea
Merge branch 'main' into dev
2025-08-19 13:51:02 +03:00
kijai
5a8f8730bd
Fix for using additional scaled model and merging loras
2025-08-19 13:31:56 +03:00
kijai
7fc5ec305d
Refactor model loading
2025-08-16 16:03:15 +03:00
kijai
43447dbaf9
Update utils.py
2025-08-14 16:57:01 +03:00
kijai
d446e97309
Experimental RAAG (Ratio Aware Adaptive Guidance) implementation
...
https://arxiv.org/abs/2508.03442
Available in experimental_args, alpha > 0 = enabled, default value is 1.0
2025-08-08 00:17:32 +03:00
kabachuha
fd285a7e7d
implement tangential guidance
2025-07-28 21:08:10 +03:00
kijai
998a69cc0a
Update utils.py
2025-07-28 10:46:46 +03:00
kijai
3567202e75
Fix compile and support radial attn in the DF sampler
2025-07-27 14:19:31 +03:00
kijai
35e637cbfd
Allow merging LoRA to fp8 scaled models
2025-07-24 11:51:36 +03:00
kijai
5e56e3649f
fix control lora
2025-07-23 16:20:30 +03:00
kijai
296baa30ce
Add WanVideoSetLoRAs
...
Node to set the LoRA weights to use with the unmerged LoRA mode, not able to merge LoRAs but allows instant LoRA switching without any loading times. The effect of unmerged LoRAs is stronger and differs from merged LoRAs.
2025-07-21 22:39:17 +03:00
kijai
4bf78e5316
Possibly reduce LoRA loading memory usage
2025-07-21 15:25:07 +03:00
kijai
768a4b2d90
Fix radial attention cache not resetting on resolution change, add more robust supported dimension check
2025-07-20 12:15:01 +03:00
kijai
c08296747f
more refactoring
2025-07-18 16:31:34 +03:00
kijai
e011d031a8
Merge branch 'pr/799'
2025-07-17 19:33:10 +03:00
kijai
d8e90a94c1
more advanced input for Pusa
2025-07-17 19:10:24 +03:00
Alexander Measure
88f50a8fd9
Update apply lora low_mem_load
...
Avoid creating 2 copies of WAN in VRAM when applying a LoRA by setting inplace_update=True on model.patch_weight_to_device.
2025-07-14 14:10:56 -04:00
kijai
127bfa2c99
refactor self attention
2025-07-11 23:00:59 +03:00
kijai
a57d7aa002
lora loading progressbar
2025-07-08 09:27:58 +03:00
kijai
ed068e1fe3
Fix lowvram lora load when not using quantization
2025-04-29 02:21:11 +03:00
kijai
7d005201a2
Allow loading UniAnimate LoRA with low vram load as well
2025-04-20 18:05:43 +03:00
kijai
7462356743
Support UniAnimate-DiT
...
https://github.com/ali-vilab/UniAnimate-DiT
2025-04-18 19:43:40 +03:00
kijai
11e9166d0b
Fix low vram lora load
2025-04-10 18:04:17 +03:00
kijai
c3e96fd6f9
Experimental clip embed enhancement through tiling
...
Allows more detail in the clip embeds as well as details from non square images without stretching
Idea and most of the code is from Matteo:
https://github.com/cubiq/ComfyUI_IPAdapter_plus/blob/9d076a3df0d2763cef5510ec5ab807f6632c39f5/utils.py#L181
2025-03-21 19:41:49 +02:00
kijai
635001f3e8
Fix torch.compile with lora reloading in control workflows
2025-03-19 22:47:53 +02:00
kijai
8ac0da07c6
Add SLG
...
Credits to AmericanPresidentJimmyCarter: https://github.com/deepbeepmeep/Wan2GP/pull/61
2025-03-13 17:29:15 +02:00
kijai
83682769d4
Add low memory LoRA loading option
2025-03-12 21:55:35 +02:00
kijai
b505075850
simplify LoRA loading further
2025-03-12 19:46:01 +02:00
kijai
d3129c6996
Add alternative offloading method from Diffsynth
2025-03-02 14:37:08 +02:00
kijai
55dcc33d00
Add block swap debug prints and cleanup unused code
2025-03-02 10:49:46 +02:00
kijai
2ee9543d48
init
2025-02-25 20:03:08 +02:00
kijai