sd: adjust VAE tile size according to sdtiledvae#2208
Draft
wbruna wants to merge 1 commit into
Draft
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
I forgot to send this one. The idea is changing the VAE tile size to use a specific fraction of the non-tiled VAE image limit: larger tiles tend to be faster, and avoid artifacts (since the model has more image context to work with).
For instance, in the default SDXL config (768x768), it increases the tile size from 32x32 to 48x48, which for SDXL resolutions is ~30% faster.
In principle, we could use any ratio, but values close to 1 are problematic because images a bit above the threshold could take almost four times longer than right at the threshold (non-tiles -> 4 tiles for a square image). In my tests, 2/3 worked fine, too; the current 1/2 is a bit more conservative.
Leaving it as draft for now, because I know we are close to the next release, and this isn't exactly a priority.