Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Merging the latest master changes from the original repo #3

Draft
wants to merge 712 commits into
base: master
Choose a base branch
from
Draft

Merging the latest master changes from the original repo #3

wants to merge 712 commits into from

Conversation

ghost
Copy link

@ghost ghost commented Jul 17, 2024

Merging the latest master changes from the original repo

@ghost
Copy link
Author

ghost commented Jul 17, 2024

@fksato
why can't i add reviewers in this PR ?

@ghost ghost self-assigned this Jul 23, 2024
@ghost ghost marked this pull request as draft July 23, 2024 09:18
huchenlei and others added 27 commits November 15, 2024 20:17
@huchenlei
Update web content to release v1.3.44 (#5620)
4ac401a 
* Update web content to release v1.3.44

* nit
@huchenlei
Fix 3.8 compatibility in user_manager.py (#5645)
22a1d7c
@comfyanonymous
Add transformer options blocks replace patch to mochi.
41886af
@comfyanonymous
Support block replace patches in auraflow.
d9f9096
@comfyanonymous
Add a skip layer guidance node that can also skip single layers.
9a0a5d3 
This one should work for skipping the single layers of models like Flux
and Auraflow.

If you want to see how these models work and how many double/single layers
they have see the "ModelMerge*" nodes for the specific model.
@huchenlei
Update UI screenshot in README (#5666)
9cc90ee 
* Update UI ScreenShot in README

* Remove legacy UI screenshot file

* nit

* nit
@comfyanonymous
Refactor inpaint/ip2p code.
b699a15
@yoland68
Add terminal size fallback (#5623)
f498d85
@comfyanonymous
Add boolean to InpaintModelConditioning to disable the noise mask.
156a287
@comfyanonymous
Rename add_noise_mask -> noise_mask.
8986151
@comfyanonymous
Skip layer guidance now works on stable audio model.
22535d0
@comfyanonymous
Fix mask issue with attention_xformers.
07f6eea
@comfyanonymous
Update readme.
772e620
@comfyanonymous
Add some new weight patching functionality.
41444b5 
Add a way to reshape lora weights.

Allow weight patches to all weight not just .weight and .bias

Add a way for a lora to set a weight to a specific value.
@comfyanonymous
Support new flux model variants.
8f0009a
@comfyanonymous
Fix mask issue in some attention functions.
2fd9c13
@comfyanonymous
Support Lightricks LTX-Video model.
5e16f1d
@comfyanonymous
Add LTX-Video support to the Readme.
0b734de
@comfyanonymous
Remove print.
5818f6c
@comfyanonymous
Remove prints.
94323a2
@comfyanonymous
Some fixes to the lowvram system.
bc6be6c
@comfyanonymous
LTXV lowvram fixes.
e5c3f4b
@comfyanonymous
Fix some tiled VAE decoding issues with LTX-Video.
6e8cdcd
@comfyanonymous
Some improvements to the lowvram unloading.
839ed33
@comfyanonymous
Skip layer guidance node now works on LTX-Video.
ab885b3
@spacepxl
set LTX min length to 1 for t2i (#5750)
7126ecf 
At length=1, the LTX model can do txt2img and img2img with no other changes required.
@40476
Update README.md (#5707)
3d80271
comfyanonymous and others added 30 commits January 16, 2025 14:54
@comfyanonymous
Bump ComfyUI version to 0.3.11
619b8cd
@comfyanonymous
Fix cosmos VAE failing with videos longer than 121 frames.
cca96a8
@comfyanonymous
Fix some cosmos fp8 issues.
0aa2368
@comfyanonymous
Bump ComfyUI version to v0.3.12
55add50
@comfyanonymous
Add that nvidia cosmos is supported to the README.
7fc3ccd
@comfyanonymous
Add warning when using old pytorch versions.
2f3ab40
@comfyanonymous
Uni pc sampler now works with audio and video models.
507199d
@catboxanon
PromptServer: Return 400 for empty filename param (#6504)
3a3910f
@catboxanon
Remove comfy.samplers self-import (#6506)
b1a0213
@comfy-pr-bot @huchenlei
Update frontend to v1.7.14 (#6522)
b4de04a 
Co-authored-by: huchenlei <[email protected]>
@kit1980
Use torch.special.expm1 (#6388)
ebf038d 
* Use `torch.special.expm1`

This function provides greater precision than `exp(x) - 1` for small values of `x`.

Found with TorchFix https://github.com/pytorch-labs/torchfix/

* Use non-alias
@comfyanonymous
LatentBatch fix for video latents
a00e148
@comfyanonymous
Cleanup old TODO.
d8a7a32
@comfyanonymous
Add FluxDisableGuidance node to disable using the guidance embed.
fb2ad64
@comfyanonymous
Add missing case to CLIPLoader.
d303cb5
@chaObserv
Add gradient estimation sampler (#6554)
e857dd4
@comfyanonymous
Refactor and fixes for video latents.
a7fe0a9
@comfyanonymous
Remove support for python 3.8.
d6bbe8c
@huchenlei
[i18n] Add /i18n endpoint to provide all custom node translations (#6558
a058f52 
)

* [i18n] Add /i18n endpoint to provide all custom node translations

* Sort glob result for deterministic ordering

* Update comment
@huchenlei
Add utils/ to web server developer codeowner (#6570)
ca69b41
@jtydhr88
remove some params from load 3d node (#6436)
f3566f0
@huchenlei
Remove unused function lcm in conds.py (#6572)
dfa2b6d
@comfyanonymous
Remove useless code.
96e2a45
@webfiltered
Remove redundant code (#6576)
ce557cf
@comfyanonymous
Remove useless code.
14ca5f5
@comfyanonymous
Update nightly pytorch ROCm command in Readme.
7fbf4b7
@comfyanonymous
Print ComfyUI version.
6d21740
@comfyanonymous
Remove redundant code.
67feb05
@comfyanonymous
Better CLIPTextEncode error when clip input is None.
4f011b9
@comfyanonymous
Lower minimum ratio of loaded weights on Nvidia.
255edf2
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.