feat: add KL Optimal scheduler #1098
Open
+47
−6
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
Contributor
|
LGTM, and seems to be working fine.
I don't know if that order matters much. When I implemented the LCM scheduler, I simply placed it at the end of the list. |
daniandtheweb
force-pushed
the
kl_optimal
branch
from
5ea6262 to
951043f
Compare
Contributor
Author
|
I was actually thinking about the order of the schedulers based on importance: of course the first would be discrete and karras as they're historically the most used and last should be LCM as it's more like a niche scheduler for a single specific application. I'm really not sure where kl_optimal should be placed compared to all the others, but if it's okay like this I think the PR is ready. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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.
This PR adds support for the KL Optimal scheduler.
I've taken inspiration from the Python implementation in ComfyUI and stable-diffusion-webui.
The resulting images seem correct. If the positioning of the scheduler looks good (after smoothstep and before lcm), I think this could be ready to review.