kijai
6bc53b771d
refactor scheduler import
2025-07-17 15:56:55 +03:00
kijai
2263d02a0a
basic Pusa support
2025-07-17 10:06:01 +03:00
kijai
17d48e3e45
rename VACE Model - VACE Module for clarity
2025-07-14 17:38:08 +03:00
kijai
a37c0e6ac8
Update nodes.py
2025-07-14 17:28:17 +03:00
kijai
24d3de2126
Add EasyCache
2025-07-14 17:27:10 +03:00
kijai
e1f5d185ee
Use original Multitalk attention code with exactly 2 speakers
2025-07-14 17:06:48 +03:00
kijai
f96128bf2c
Make multitalk sampling actually use the seed...
2025-07-14 09:20:03 +03:00
kijai
8fd62b3e78
Multitalk wav2vec model doesn't actually need the .pt file
...
chinese-wav2vec2-base-fairseq-ckpt.pt can be safely deleted and won't be autodownloaded in the future
2025-07-14 09:13:41 +03:00
kijai
08669de287
Fix FantasyTalking when wav2vec loaded on offload_device
2025-07-13 23:52:19 +03:00
kijai
b6d156afa5
Update nodes.py
2025-07-13 19:13:20 +03:00
kijai
1e4ed578ba
Update pyproject.toml
2025-07-13 18:42:17 +03:00
kijai
365a1ce1d6
more chunking and refactoring
2025-07-12 20:18:52 +03:00
kijai
895c92febd
Update model.py
2025-07-12 03:04:46 +03:00
kijai
039b211ef7
cleanup
2025-07-12 02:06:02 +03:00
kijai
127bfa2c99
refactor self attention
2025-07-11 23:00:59 +03:00
kijai
880194a1f9
bump version: 1.2.3
2025-07-11 01:13:25 +03:00
kijai
98395f94ce
Update nodes_model_loading.py
2025-07-11 01:11:43 +03:00
kijai
1077d2323c
fix max res selection
2025-07-10 23:41:54 +03:00
kijai
15f0e62e5f
Add chunked RoPE option to reduce peak VRAM usage when not using torch.compile
2025-07-10 17:15:06 +03:00
kijai
49de5335c0
fix multi lora loader
2025-07-10 16:30:20 +03:00
kijai
a9ba0e1cb4
small possible optimizations
2025-07-10 14:18:04 +03:00
kijai
5df3154112
Update requirements.txt
2025-07-10 12:53:37 +03:00
kijai
6e930d16cc
Update nodes.py
2025-07-10 12:41:40 +03:00
kijai
d71bed76af
display possible lora metadata
2025-07-10 12:33:40 +03:00
kijai
da2574d1bb
WanVideoVACEStartToEndFrame fixes
2025-07-09 16:17:01 +03:00
kijai
456d04e318
fix vid2vid
2025-07-08 19:14:01 +03:00
kijai
b65f161e8b
Fix gguf + lora alpha scaling
2025-07-08 16:47:41 +03:00
kijai
ce0bae1839
revert this
2025-07-08 16:23:29 +03:00
kijai
948805b6e5
remove print
2025-07-08 16:22:43 +03:00
kijai and kabachuha
da98636599
FreeInit
...
https: //github.com/TianxingWu/FreeInit
Co-Authored-By: kabachuha <14872007+kabachuha@users.noreply.github.com >
2025-07-08 15:09:54 +03:00
kijai
8500514ef9
gguf + lora application fix
2025-07-08 12:53:29 +03:00
kijai
20e9914645
Update gguf.py
2025-07-08 12:33:50 +03:00
kijai
372e33bd5b
possible fix for some gguf + torch.compile issues
2025-07-08 11:32:27 +03:00
kijai
a57d7aa002
lora loading progressbar
2025-07-08 09:27:58 +03:00
kijai
d16e2aa2ff
Apply possible LoRA alpha with GGUF too
2025-07-08 09:27:22 +03:00
kijai
b355f2b839
Update requirements.txt
2025-07-07 19:04:31 +03:00
Jukka Seppänen
aa6fc5b12e
Update readme.md
2025-07-07 17:22:06 +03:00
kijai
53ac085fc0
fix loop decode
2025-07-07 16:57:19 +03:00
kijai
5e1a5c6cf6
these won't work together
2025-07-07 01:44:06 +03:00
kijai
79ed360009
fix Q5 etc.
2025-07-07 01:31:27 +03:00
kijai
ea183da9ba
bump version
2025-07-06 23:58:22 +03:00
kijai
d903ea7d3d
small fixes, add comfy pbar for model loading
2025-07-06 23:55:37 +03:00
kijai
a35142b006
Update nodes_model_loading.py
2025-07-06 23:20:04 +03:00
kijai
6ae5531bad
Basic GGUF support (yes, really)
2025-07-06 21:37:45 +03:00
kijai
ee1250c6bb
fix cfg multitalk
2025-07-06 01:04:17 +03:00
Jukka Seppänen
20af8955c3
Merge pull request #687 from peteromallet/main
...
feat: Add ExtractStartFramesForContinuations node
2025-07-04 20:04:08 +03:00
kijai
79134368b4
bump version
2025-07-04 20:00:43 +03:00
kijai
78cc3050e0
update FLF2V example
2025-07-04 20:00:21 +03:00
kijai
490b0a8ead
better context window progress bar, add VAE decode progress bar
2025-07-04 20:00:03 +03:00
kijai
32e5c4d200
update examples
2025-07-04 17:19:51 +03:00
kijai
3d7801cee4
Update nodes.py
2025-07-04 16:57:20 +03:00
kijai
ca3c58bb1a
Merge branch 'main' into multitalk
2025-07-04 16:43:53 +03:00
kijai
811dd32713
Update wanvideo_ATI_testing_01.json
2025-07-04 16:43:19 +03:00
kijai
74c6d25486
update WanVideoVACEStartToEndFrame
2025-07-04 16:43:13 +03:00
kijai
974dd656da
some offloading for multitalk
2025-07-03 21:56:19 +03:00
kijai
4e6a85a148
Merge branch 'main' into multitalk
2025-07-03 21:56:06 +03:00
kijai
70e3c6b524
update
2025-07-03 21:41:44 +03:00
kijai
eb6de2e8d0
Update nodes.py
2025-07-03 19:18:28 +03:00
kijai and Rudra-ai-coder
475f371016
Multiple talkers
...
Initial commit, works but needs more utility for the mask creation.
Based mostly on Rudra-ai-coder's modifications.
Co-Authored-By: Rudra-ai-coder <177262225+rudra-ai-coder@users.noreply.github.com >
2025-07-03 19:03:12 +03:00
kijai
06b932792f
Update nodes.py
2025-07-02 17:30:35 +03:00
kijai
74a441ef93
Update nodes.py
2025-07-02 17:24:57 +03:00
kijai
de279dd060
update
2025-07-02 17:15:18 +03:00
kijai
ebe3068a8d
progress bar
2025-07-02 16:59:45 +03:00
kijai
8e9fb0e11c
Update nodes.py
2025-07-02 16:47:19 +03:00
kijai
d687833c15
Create wanvideo_multitalk_test_02.json
2025-07-02 16:47:15 +03:00
kijai
0a11c67a0c
MultiTalk sampling
...
New node (WanVideoImageToVideoMultiTalk) that also enables the continuous sampling method from the original code. Compared to context windows this works better for shorter clips, but degrades longer it goes.
2025-07-02 16:18:04 +03:00
kijai
49430f900b
Fix for transformers 4.53.0
2025-07-01 12:00:28 +03:00
kijai
f621d2a5fc
Update model.py
2025-06-30 20:31:39 +03:00
kijai
8479624614
fix split prompting
2025-06-20 17:31:01 +03:00
kijai
2da36b7fda
context window progress bar
...
better than nothing, and allow cancelling mid context
2025-06-20 17:29:27 +03:00
kijai
b605308687
Update nodes.py
2025-06-20 16:33:55 +03:00
kijai
547bd4e45a
Update nodes.py
2025-06-20 16:18:07 +03:00
kijai
4334ed58b4
Update nodes.py
2025-06-20 15:26:13 +03:00
kijai
c44eca13a9
Update nodes.py
2025-06-20 15:22:25 +03:00
kijai
c3ab39b068
Update requirements.txt
2025-06-20 15:18:13 +03:00
kijai
157fbb1b7f
Merge branch 'main' into multitalk
2025-06-20 13:53:31 +03:00
kijai
2b96d9cef2
WanVideoLoraSelectMulti
2025-06-20 13:53:13 +03:00
kijai
8387a64b15
bump version
2025-06-20 10:57:13 +03:00
kijai
38c98a6476
Create wanvideo_multitalk_test_01.json
2025-06-20 01:08:16 +03:00
kijai
263333a0d1
tiny vae adjustment
2025-06-19 20:38:23 +03:00
kijai
09b4c3a865
support context windows, torch compile fixes
2025-06-19 20:29:32 +03:00
kijai
f3614e6720
Update nodes.py
2025-06-19 00:56:36 +03:00
kijai
96e8914b33
Fix audio cfg
2025-06-18 21:19:24 +03:00
kijai
1fe72d27aa
loudness norm
2025-06-18 18:35:06 +03:00
kijai
b5cae8bf1a
Fix FLF2V model
2025-06-18 15:49:01 +03:00
kijai
58104b620f
init
2025-06-18 15:41:53 +03:00
Jukka Seppänen
058286fc0f
Merge pull request #657 from ObiLeek/main
...
refactor(decode): Replace min-max scaling with standard [-1, 1] conversion
2025-06-17 15:14:03 +03:00
kijai
dd1f3c3ecb
fix
2025-06-17 12:14:56 +03:00
kijai
b322192d93
Support minimaxremover
...
https://minimax-remover.github.io/
2025-06-17 09:45:12 +03:00
kijai
c3ee35f3ec
Update basic_flowmatch.py
2025-06-13 13:06:12 +03:00
kijai
ad43eed0d0
Add MagCache and consolidate cache_args to support both
...
Not really getting great results yet with MagCache, but at least pretty much on bar with TeaCache so it can be an option that hopefully improves in time
2025-06-13 12:31:13 +03:00
kijai
e7c39757f6
Allow compiling crossattn
...
I don't remember why I even disabled this... seems to work, big effect with NAG
2025-06-12 19:04:15 +03:00
kijai
f6df3d75ef
Merge branch 'pr/643'
2025-06-12 18:32:20 +03:00
kijai
34fecc3063
Update nodes.py
2025-06-12 18:19:52 +03:00
kijai
d34cabc985
Cleanup, create own node for NAG
2025-06-12 18:18:28 +03:00
kijai
23796233df
Some fixes and adjustments
2025-06-12 17:26:54 +03:00
kijai
bac29263da
Update nodes.py
2025-06-12 16:39:36 +03:00
kijai
ced1ddaa1a
Update basic_flowmatch.py
2025-06-12 16:29:53 +03:00
kijai
5917f51837
Small adjustments and options to dwpose detection
2025-06-12 16:29:31 +03:00
kijai
6eddec54a6
fix
2025-06-09 23:18:01 +03:00
kijai