-
-
Notifications
You must be signed in to change notification settings - Fork 545
Beta feat: Cfg skimming #1122
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.
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
Open
Mu5hr00moO
wants to merge
40
commits into
kijai:main
Choose a base branch
from
Mu5hr00moO:CFG_Skimming
base: main
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Open
Beta feat: Cfg skimming #1122
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Author
… helper node with scheduler list
FIX: forgot to remove one import
Author
|
Code had wrong indentations in CFG part - fixed now. Here some samples with base settings:
1: Comparison_1.mp42: Comparison_2.mp43: Comparison_3.mp4 |
Refactors the existing CFG Skimming logic into a new, modular architecture and introduces four new advanced CFG modification algorithms. The new design is built around a central controller node () which automatically detects the active algorithm based on the connected parameter node. This decouples the sampler logic from the UI and makes the system highly extensible. Implementation of code based on Skimmed_CFG by Extraltodeus (https://github.com/Extraltodeus/Skimmed_CFG). The existing Skimmed_CFG methods (, , etc.) are now managed by a dedicated node. New advanced algorithms were added: - AMI-CFG (Attention-Masked Interpolation) - PAG-CFG (Perceptual Adaptive Guidance) - SSDT-CFG (Scheduled Skimming w/ Dynamic Thresholding) - LSS-CFG (Latent Space Smoothing) All nodes include detailed descriptions and tooltips for improved usability. The sampler logic has been simplified, delegating all CFG processing to a central dispatcher function.
cfg_args fix Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
fix Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
fix again...
…g algorithms. - Changed WanVideoScheduler and added WanVideoSchedulerPlotSettings - now supports boundary sigma split + minor patches - Improved logging details in processing functions for better traceability for CFG skimms - Removed logging for pos/neg length correction (works, still leaving hardcoded align switch tho) - Changed max CFG to 100 for sampler
vu0tran
reviewed
Oct 11, 2025
| from tqdm import tqdm | ||
| import inspect | ||
| import copy | ||
| from PIL import Image |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
needs a comma
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.

implementation of code based on Skimmed_CFG by Extraltodeus (https://github.com/Extraltodeus/Skimmed_CFG)
tested t2v - works
allow to use cfg with lighting loras both on high and low without burning
need moar testing tho - for now i manage to keep motion dynamics with total 4 steps (more ofc is better)