Rando717
e81bb5b70f
Update comfyui.bat
...
inline with comfyui-n.bat, but without build tools check.
2025-09-23 05:38:07 +02:00
Rando717
3d24e253b9
Update comfyui-n.bat (WMIC)
...
-removed WMIC for driver version lookup
(off by default on win11 24H2 and soon to be removed in 25H2)
-added less reliable search for AMD Software version via registry (stops at first match)
2025-09-23 05:32:50 +02:00
patientx
09055fa3ea
Merge branch 'comfyanonymous:master' into master
2025-09-22 11:57:07 +03:00
patientx
d9373263ad
Merge pull request #309 from Rando717/Rando717-deepcache
...
DeepCache removal (obsolete)
2025-09-22 11:56:44 +03:00
patientx
4173f0277a
Merge pull request #308 from Rando717/Rando717-cleanup
...
(draft) More info for end user...
2025-09-22 11:56:08 +03:00
Rando717
1cc2c89631
Update install-n.bat
2025-09-22 02:25:51 +02:00
Rando717
0fc01bff3e
Update install-legacy.bat
2025-09-22 02:25:30 +02:00
Rando717
097a865bc5
Update install-for-older-amd.bat
...
sigh...empty lines
2025-09-22 02:24:53 +02:00
Rando717
effa9afb52
Update install-n.bat
...
delete deepcache
2025-09-22 02:00:50 +02:00
Rando717
c5a3a4bc3e
Update install-legacy.bat
...
delete deepcache
2025-09-22 02:00:28 +02:00
Rando717
59079ff00a
Update install-for-older-amd.bat
...
delete deepcache
2025-09-22 01:59:46 +02:00
Rando717
1254017225
Delete deepcache-sample.json
2025-09-22 01:58:41 +02:00
comfyanonymous
27bc181c49
Set some wan nodes as no longer experimental. ( #9976 )
2025-09-21 19:48:31 -04:00
Rando717
236e103f4f
Update comfyui-n.bat
...
added line breaks for readability
2025-09-22 01:14:44 +02:00
Rando717
5dcd8d2428
add files via upload
...
uploaded nvcuda.zluda_get_nightly_flag.py to get nightly flag info inside batch
2025-09-21 19:31:18 +02:00
Rando717
25945ff604
Update comfyui-n.bat
...
-check for VS Build Tools via vswhere (assuming default path)
-get ComfyUI(batch) path, censor username if inside Users folder
-get current commit date, hash and branch
-list new commits on git pull
-get driver version via wmic
-map driver version to corresponding AMD Software (WIP)
-get [nightly build] or [release build] info for ZLUDA from nvcuda.dll via nvcuda.zluda_get_nightly_flag.py
2025-09-21 19:28:30 +02:00
patientx
9d2b926f56
Merge branch 'comfyanonymous:master' into master
2025-09-21 16:57:32 +03:00
comfyanonymous
d1d9eb94b1
Lower wan memory estimation value a bit. ( #9964 )
...
Previous pr reduced the peak memory requirement.
2025-09-20 22:09:35 -04:00
Kohaku-Blueleaf
7be2b49b6b
Fix LoRA Trainer bugs with FP8 models. ( #9854 )
...
* Fix adapter weight init
* Fix fp8 model training
* Avoid inference tensor
2025-09-20 21:24:48 -04:00
Jedrzej Kosinski
9ed3c5cc09
[Reviving #5709 ] Add strength input to Differential Diffusion ( #9957 )
...
* Update nodes_differential_diffusion.py
* Update nodes_differential_diffusion.py
* Make strength optional to avoid validation errors when loading old workflows, adjust step
---------
Co-authored-by: ThereforeGames <eric@sparknight.io>
2025-09-20 21:10:39 -04:00
patientx
945cd4af42
Merge branch 'comfyanonymous:master' into master
2025-09-20 16:17:58 +03:00
comfyanonymous
66241cef31
Add inputs for character replacement to the WanAnimateToVideo node. ( #9960 )
2025-09-20 02:24:10 -04:00
patientx
c62e820d45
Merge branch 'comfyanonymous:master' into master
2025-09-20 01:51:06 +03:00
comfyanonymous
e8df53b764
Update WanAnimateToVideo to more easily extend videos. ( #9959 )
2025-09-19 18:48:56 -04:00
Alexander Piskun
852704c81a
fix(seedream4): add flag to ignore error on partial success ( #9952 )
2025-09-19 16:04:51 -04:00
Alexander Piskun
9fdf8c25ab
api_nodes: reduce default timeout from 7 days to 2 hours ( #9918 )
2025-09-19 16:02:43 -04:00
comfyanonymous
dc95b6acc0
Basic WIP support for the wan animate model. ( #9939 )
2025-09-19 03:07:17 -04:00
Christian Byrne
711bcf33ee
Bump frontend to 1.26.13 ( #9933 )
2025-09-19 03:03:30 -04:00
comfyanonymous
24b0fce099
Do padding of audio embed in model for humo for more flexibility. ( #9935 )
2025-09-18 19:54:16 -04:00
Jodh Singh
1ea8c54064
make kernel of same type as image to avoid mismatch issues ( #9932 )
2025-09-18 19:51:16 -04:00
DELUXA
8d6653fca6
Enable fp8 ops by default on gfx1200 ( #9926 )
2025-09-18 19:50:37 -04:00
patientx
3ffaddd005
Merge pull request #307 from Rando717/Rando717-cleanup
...
removed pause on Git error
2025-09-18 03:02:55 +03:00
patientx
50e281dc6d
Merge branch 'comfyanonymous:master' into master
2025-09-18 03:02:08 +03:00
comfyanonymous
dd611a7700
Support the HuMo 17B model. ( #9912 )
2025-09-17 18:39:24 -04:00
Rando717
685b542d17
Update comfyui.bat (git)
...
- removed pause on git check, so batch can continue on error
- expanded error message with Git url
2025-09-17 23:51:38 +02:00
Rando717
005853ccaf
Update comfyui-n.bat (git)
2025-09-17 23:49:49 +02:00
patientx
a8b63b21fe
Merge branch 'comfyanonymous:master' into master
2025-09-17 12:29:45 +03:00
comfyanonymous
9288c78fc5
Support the HuMo model. ( #9903 )
2025-09-17 00:12:48 -04:00
rattus128
e42682b24e
Reduce Peak WAN inference VRAM usage ( #9898 )
...
* flux: Do the xq and xk ropes one at a time
This was doing independendent interleaved tensor math on the q and k
tensors, leading to the holding of more than the minimum intermediates
in VRAM. On a bad day, it would VRAM OOM on xk intermediates.
Do everything q and then everything k, so torch can garbage collect
all of qs intermediates before k allocates its intermediates.
This reduces peak VRAM usage for some WAN2.2 inferences (at least).
* wan: Optimize qkv intermediates on attention
As commented. The former logic computed independent pieces of QKV in
parallel which help more inference intermediates in VRAM spiking
VRAM usage. Fully roping Q and garbage collecting the intermediates
before touching K reduces the peak inference VRAM usage.
2025-09-16 19:21:14 -04:00
patientx
9cdd9e38d2
Merge branch 'comfyanonymous:master' into master
2025-09-17 00:30:44 +03:00
comfyanonymous
a39ac59c3e
Add encoder part of whisper large v3 as an audio encoder model. ( #9894 )
...
Not useful yet but some models use it.
2025-09-16 01:19:50 -04:00
blepping
1a85483da1
Fix depending on asserts to raise an exception in BatchedBrownianTree and Flash attn module ( #9884 )
...
Correctly handle the case where w0 is passed by kwargs in BatchedBrownianTree
2025-09-15 20:05:03 -04:00
patientx
a08c1e1613
Merge branch 'comfyanonymous:master' into master
2025-09-16 01:22:18 +03:00
comfyanonymous
47a9cde5d3
Support the omnigen2 umo lora. ( #9886 )
2025-09-15 18:10:55 -04:00
patientx
31ed58199b
Merge pull request #304 from Rando717/Rando717-cleanup
...
comfyui.bat cleanup with Git check
2025-09-15 12:35:24 +03:00
Rando717
d6495b10a7
Update comfyui.bat (typo)
2025-09-15 00:49:28 +02:00
Rando717
6571c9f3ef
Update comfyui.bat
2025-09-14 23:39:19 +02:00
Rando717
2f256e48cb
Update comfyui-n.bat
...
- reverted previous commit (removes colored + - signs)
- added check for git (install+version) and current commit hash
- stylized output: [INFO], [PASS], [FAIL]
- added launching app info
2025-09-14 23:25:39 +02:00
patientx
b01a6f3baf
Update README.md
2025-09-14 13:24:24 +03:00
patientx
09bd37e843
Merge branch 'comfyanonymous:master' into master
2025-09-14 13:23:58 +03:00
