patientx
7ff01ded58
Merge branch 'comfyanonymous:master' into master
2025-08-21 09:24:26 +03:00
comfyanonymous
0963493a9c
Support for Qwen Diffsynth Controlnets canny and depth. ( #9465 )
...
These are not real controlnets but actually a patch on the model so they
will be treated as such.
Put them in the models/model_patches/ folder.
Use the new ModelPatchLoader and QwenImageDiffsynthControlnet nodes.
2025-08-20 22:26:37 -04:00
patientx
6dca25e2a8
Merge branch 'comfyanonymous:master' into master
2025-08-20 10:14:34 +03:00
comfyanonymous
8d38ea3bbf
Fix bf16 precision issue with qwen image embeddings. ( #9441 )
2025-08-20 02:58:54 -04:00
comfyanonymous
5a8f502db5
Disable prompt weights for qwen. ( #9438 )
2025-08-20 01:08:11 -04:00
comfyanonymous
7cd2c4bd6a
Qwen rotary embeddings should now match reference code. ( #9437 )
2025-08-20 00:45:27 -04:00
comfyanonymous
dfa791eb4b
Rope fix for qwen vl. ( #9435 )
2025-08-19 20:47:42 -04:00
patientx
1cbb5fdc14
Merge branch 'comfyanonymous:master' into master
2025-08-19 10:21:12 +03:00
comfyanonymous
4977f203fa
P2 of qwen edit model. ( #9412 )
...
* P2 of qwen edit model.
* Typo.
* Fix normal qwen.
* Fix.
* Make the TextEncodeQwenImageEdit also set the ref latent.
If you don't want it to set the ref latent and want to use the
ReferenceLatent node with your custom latent instead just disconnect the
VAE.
2025-08-18 22:38:34 -04:00
patientx
3f09b4dba5
Merge branch 'comfyanonymous:master' into master
2025-08-18 15:14:34 +03:00
Jedrzej Kosinski
7f3b9b16c6
Make step index detection much more robust ( #9392 )
2025-08-17 18:54:07 -04:00
comfyanonymous
ed43784b0d
WIP Qwen edit model: The diffusion model part. ( #9383 )
2025-08-17 16:45:39 -04:00
patientx
64d6cf045e
Merge branch 'comfyanonymous:master' into master
2025-08-17 11:29:13 +03:00
comfyanonymous
0f2b8525bc
Qwen image model refactor. ( #9375 )
2025-08-16 17:51:28 -04:00
patientx
5a21015adb
Merge branch 'comfyanonymous:master' into master
2025-08-16 09:54:01 +03:00
comfyanonymous
1702e6df16
Implement wan2.2 camera model. ( #9357 )
...
Use the old WanCameraImageToVideo node.
2025-08-15 17:29:58 -04:00
patientx
eb283b5fd7
Merge branch 'comfyanonymous:master' into master
2025-08-16 00:26:31 +03:00
comfyanonymous
c308a8840a
Add FluxKontextMultiReferenceLatentMethod node. ( #9356 )
...
This node is only useful if someone trains the kontext model to properly
use multiple reference images via the index method.
The default is the offset method which feeds the multiple images like if
they were stitched together as one. This method works with the current
flux kontext model.
2025-08-15 15:50:39 -04:00
patientx
13f5f9d78f
Merge branch 'comfyanonymous:master' into master
2025-08-15 10:54:10 +03:00
comfyanonymous
e08ecfbd8a
Add warning when using old pytorch. ( #9347 )
2025-08-15 00:22:26 -04:00
comfyanonymous
4e5c230f6a
Fix last commit not working on older pytorch. ( #9346 )
2025-08-14 23:44:02 -04:00
Xiangxi Guo (Ryan)
f0d5d0111f
Avoid torch compile graphbreak for older pytorch versions ( #9344 )
...
Turns out torch.compile has some gaps in context manager decorator
syntax support. I've sent patches to fix that in PyTorch, but it won't
be available for all the folks running older versions of PyTorch, hence
this trivial patch.
2025-08-14 23:41:37 -04:00
comfyanonymous
ad19a069f6
Make SLG nodes work on Qwen Image model. ( #9345 )
2025-08-14 23:16:01 -04:00
patientx
a927fbd99b
Merge branch 'comfyanonymous:master' into master
2025-08-14 12:16:50 +03:00
Jedrzej Kosinski
e4f7ea105f
Added context window support to core sampling code ( #9238 )
...
* Added initial support for basic context windows - in progress
* Add prepare_sampling wrapper for context window to more accurately estimate latent memory requirements, fixed merging wrappers/callbacks dicts in prepare_model_patcher
* Made context windows compatible with different dimensions; works for WAN, but results are bad
* Fix comfy.patcher_extension.merge_nested_dicts calls in prepare_model_patcher in sampler_helpers.py
* Considering adding some callbacks to context window code to allow extensions of behavior without the need to rewrite code
* Made dim slicing cleaner
* Add Wan Context WIndows node for testing
* Made context schedule and fuse method functions be stored on the handler instead of needing to be registered in core code to be found
* Moved some code around between node_context_windows.py and context_windows.py
* Change manual context window nodes names/ids
* Added callbacks to IndexListContexHandler
* Adjusted default values for context_length and context_overlap, made schema.inputs definition for WAN Context Windows less annoying
* Make get_resized_cond more robust for various dim sizes
* Fix typo
* Another small fix
2025-08-13 21:33:05 -04:00
Simon Lui
c991a5da65
Fix XPU iGPU regressions ( #9322 )
...
* Change bf16 check and switch non-blocking to off default with option to force to regain speed on certain classes of iGPUs and refactor xpu check.
* Turn non_blocking off by default for xpu.
* Update README.md for Intel GPUs.
2025-08-13 19:13:35 -04:00
patientx
804c7097fa
Merge branch 'comfyanonymous:master' into master
2025-08-13 23:56:43 +03:00
comfyanonymous
9df8792d4b
Make last PR not crash comfy on old pytorch. ( #9324 )
2025-08-13 15:12:41 -04:00
contentis
3da5a07510
SDPA backend priority ( #9299 )
2025-08-13 14:53:27 -04:00
patientx
bcafc3f7a3
Merge branch 'comfyanonymous:master' into master
2025-08-13 10:36:20 +03:00
comfyanonymous
560d38f34c
Wan2.2 fun control support. ( #9292 )
2025-08-12 23:26:33 -04:00
patientx
f80a9bb674
Merge branch 'comfyanonymous:master' into master
2025-08-12 00:33:53 +03:00
PsychoLogicAu
2208aa616d
Support SimpleTuner lycoris lora for Qwen-Image ( #9280 )
2025-08-11 16:56:16 -04:00
patientx
c2686a3968
Merge branch 'comfyanonymous:master' into master
2025-08-10 12:09:19 +03:00
comfyanonymous
5828607ccf
Not sure if AMD actually support fp16 acc but it doesn't crash. ( #9258 )
2025-08-09 12:49:25 -04:00
patientx
89499c6fae
Merge branch 'comfyanonymous:master' into master
2025-08-08 11:40:07 +03:00
comfyanonymous
735bb4bdb1
Users report gfx1201 is buggy on flux with pytorch attention. ( #9244 )
2025-08-08 04:21:00 -04:00
patientx
8795ae98aa
Merge branch 'comfyanonymous:master' into master
2025-08-06 20:24:47 +03:00
flybirdxx
4c3e57b0ae
Fixed an issue where qwenLora could not be loaded properly. ( #9208 )
2025-08-06 13:23:11 -04:00
patientx
2e39e0999f
Update zluda.py
2025-08-05 19:21:20 +03:00
patientx
28957a7bd6
Merge branch 'comfyanonymous:master' into master
2025-08-05 13:37:09 +03:00
comfyanonymous
d044a24398
Fix default shift and any latent size for qwen image model. ( #9186 )
2025-08-05 06:12:27 -04:00
patientx
e419bade03
Merge pull request #244 from sfinktah/sfink-zluda-is-nasty
...
Bad ideas from zluda update.
2025-08-05 09:48:53 +03:00
patientx
ea8122f065
Merge branch 'comfyanonymous:master' into master
2025-08-05 09:47:31 +03:00
comfyanonymous
c012400240
Initial support for qwen image model. ( #9179 )
2025-08-04 22:53:25 -04:00
Christopher Anderson
4f853403fe
Bad ideas from zluda update.
2025-08-05 06:00:55 +10:00
patientx
88b7fe87ff
Merge branch 'comfyanonymous:master' into master
2025-08-04 12:38:56 +03:00
comfyanonymous
03895dea7c
Fix another issue with the PR. ( #9170 )
2025-08-04 04:33:04 -04:00
comfyanonymous
84f9759424
Add some warnings and prevent crash when cond devices don't match. ( #9169 )
2025-08-04 04:20:12 -04:00
comfyanonymous
7991341e89
Various fixes for broken things from earlier PR. ( #9168 )
2025-08-04 04:02:40 -04:00
patientx
37415c40c1
device identification and setting triton arch override
2025-08-04 10:44:18 +03:00
patientx
d823c0c615
Merge branch 'comfyanonymous:master' into master
2025-08-04 10:42:15 +03:00
comfyanonymous
140ffc7fdc
Fix broken controlnet from last PR. ( #9167 )
2025-08-04 03:28:12 -04:00
comfyanonymous
182f90b5ec
Lower cond vram use by casting at the same time as device transfer. ( #9159 )
2025-08-04 03:11:53 -04:00
patientx
7258461c23
Merge branch 'comfyanonymous:master' into master
2025-08-03 16:33:54 +03:00
comfyanonymous
aebac22193
Cleanup. ( #9160 )
2025-08-03 07:08:11 -04:00
patientx
da4fc8189a
Merge branch 'comfyanonymous:master' into master
2025-08-03 00:17:56 +03:00
comfyanonymous
13aaa66ec2
Make sure context is on the right device. ( #9154 )
2025-08-02 15:09:23 -04:00
comfyanonymous
5f582a9757
Make sure all the conds are on the right device. ( #9151 )
2025-08-02 15:00:13 -04:00
patientx
83dbd68651
Merge branch 'comfyanonymous:master' into master
2025-08-01 14:42:25 +03:00
comfyanonymous
1e638a140b
Tiny wan vae optimizations. ( #9136 )
2025-08-01 05:25:38 -04:00
patientx
321d683af0
Merge branch 'comfyanonymous:master' into master
2025-07-31 14:49:33 +03:00
chaObserv
61b08d4ba6
Replace manual x * sigmoid(x) with torch silu in VAE nonlinearity ( #9057 )
2025-07-30 19:25:56 -04:00
comfyanonymous
da9dab7edd
Small wan camera memory optimization. ( #9111 )
2025-07-30 05:55:26 -04:00
patientx
1bd4b6489e
Merge branch 'comfyanonymous:master' into master
2025-07-30 11:11:46 +03:00
comfyanonymous
dca6bdd4fa
Make wan2.2 5B i2v take a lot less memory. ( #9102 )
2025-07-29 19:44:18 -04:00
patientx
d8ca8134c3
Merge branch 'comfyanonymous:master' into master
2025-07-29 11:56:59 +03:00
comfyanonymous
7d593baf91
Extra reserved vram on large cards on windows. ( #9093 )
2025-07-29 04:07:45 -04:00
patientx
fc4e82537c
Merge pull request #233 from sfinktah/sfink-flash-attn-gfx-startswith
...
This will allow much better support for gfx1032 and other things not …
2025-07-28 23:12:38 +03:00
patientx
7ba2a8d3b0
Merge branch 'comfyanonymous:master' into master
2025-07-28 22:15:10 +03:00
comfyanonymous
c60dc4177c
Remove unecessary clones in the wan2.2 VAE. ( #9083 )
2025-07-28 14:48:19 -04:00
Christopher Anderson
b5ede18481
This will allow much better support for gfx1032 and other things not specifically named
2025-07-29 04:21:45 +10:00
patientx
769ab3bd25
Merge branch 'comfyanonymous:master' into master
2025-07-28 15:21:30 +03:00
comfyanonymous
a88788dce6
Wan 2.2 support. ( #9080 )
2025-07-28 08:00:23 -04:00
patientx
5a45e12b61
Merge branch 'comfyanonymous:master' into master
2025-07-26 14:09:19 +03:00
comfyanonymous
0621d73a9c
Remove useless code. ( #9059 )
2025-07-26 04:44:19 -04:00
comfyanonymous
e6e5d33b35
Remove useless code. ( #9041 )
...
This is only needed on old pytorch 2.0 and older.
2025-07-25 04:58:28 -04:00
patientx
c3bf1d95e2
Merge branch 'comfyanonymous:master' into master
2025-07-25 10:20:29 +03:00
Eugene Fairley
4293e4da21
Add WAN ATI support ( #8874 )
...
* Add WAN ATI support
* Fixes
* Fix length
* Remove extra functions
* Fix
* Fix
* Ruff fix
* Remove torch.no_grad
* Add batch trajectory logic
* Scale inputs before and after motion patch
* Batch image/trajectory
* Ruff fix
* Clean up
2025-07-24 20:59:19 -04:00
patientx
970b7fb84f
Merge branch 'comfyanonymous:master' into master
2025-07-24 22:30:55 +03:00
comfyanonymous
69cb57b342
Print xpu device name. ( #9035 )
2025-07-24 15:06:25 -04:00
honglyua
0ccc88b03f
Support Iluvatar CoreX ( #8585 )
...
* Support Iluvatar CoreX
Co-authored-by: mingjiang.li <mingjiang.li@iluvatar.com>
2025-07-24 13:57:36 -04:00
patientx
30539d0d13
Merge branch 'comfyanonymous:master' into master
2025-07-24 13:59:09 +03:00
Kohaku-Blueleaf
eb2f78b4e0
[Training Node] algo support, grad acc, optional grad ckpt ( #9015 )
...
* Add factorization utils for lokr
* Add lokr train impl
* Add loha train impl
* Add adapter map for algo selection
* Add optional grad ckpt and algo selection
* Update __init__.py
* correct key name for loha
* Use custom fwd/bwd func and better init for loha
* Support gradient accumulation
* Fix bugs of loha
* use more stable init
* Add OFT training
* linting
2025-07-23 20:57:27 -04:00
chaObserv
e729a5cc11
Separate denoised and noise estimation in Euler CFG++ ( #9008 )
...
This will change their behavior with the sampling CONST type.
It also combines euler_cfg_pp and euler_ancestral_cfg_pp into one main function.
2025-07-23 19:47:05 -04:00
comfyanonymous
d3504e1778
Enable pytorch attention by default for gfx1201 on torch 2.8 ( #9029 )
2025-07-23 19:21:29 -04:00
comfyanonymous
a86a58c308
Fix xpu function not implemented p2. ( #9027 )
2025-07-23 18:18:20 -04:00
comfyanonymous
39dda1d40d
Fix xpu function not implemented. ( #9026 )
2025-07-23 18:10:59 -04:00
patientx
58f3250106
Merge branch 'comfyanonymous:master' into master
2025-07-23 23:49:46 +03:00
comfyanonymous
5ad33787de
Add default device argument. ( #9023 )
2025-07-23 14:20:49 -04:00
patientx
bd33a5d382
Merge branch 'comfyanonymous:master' into master
2025-07-23 03:27:52 +03:00
Simon Lui
255f139863
Add xpu version for async offload and some other things. ( #9004 )
2025-07-22 15:20:09 -04:00
patientx
b049c1df82
Merge branch 'comfyanonymous:master' into master
2025-07-17 00:49:42 +03:00
comfyanonymous
491fafbd64
Silence clip tokenizer warning. ( #8934 )
2025-07-16 14:42:07 -04:00
Harel Cain
9bc2798f72
LTXV VAE decoder: switch default padding mode ( #8930 )
2025-07-16 13:54:38 -04:00
patientx
d22d65cc68
Merge branch 'comfyanonymous:master' into master
2025-07-16 13:58:56 +03:00
comfyanonymous
50afba747c
Add attempt to work around the safetensors mmap issue. ( #8928 )
2025-07-16 03:42:17 -04:00
patientx
79e3b67425
Merge branch 'comfyanonymous:master' into master
2025-07-15 12:24:08 +03:00
Yoland Yan
543c24108c
Fix wrong reference bug ( #8910 )
2025-07-14 20:45:55 -04:00
patientx
3845c2ff7a
Merge branch 'comfyanonymous:master' into master
2025-07-12 14:59:05 +03:00
