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.

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

Make the default max RAM cache size more conservative #7603

Merged
merged 1 commit into from
Jan 30, 2025

Conversation

RyanJDick
Copy link
Collaborator

@RyanJDick RyanJDick commented Jan 28, 2025

Summary

This PR makes the default max RAM cache size slightly more conservative. This should help with some reports of Invoke using more RAM/VRAM than expected/desired. Of course, users always have the option to configure Invoke to use a larger cache size.

QA Instructions

The change is relatively straightforward. I tested the following cases:

  • Max RAM limit is set to 50% of available system RAM.
  • RAM limit is set to 50% of available system RAM and then reduced to 1x VRAM cache size

Reviewers are encouraged to try this out on their systems to see if default performance is worse/better/the same.

The real test will be whether this reduces the number of issue reports with the default configs.

Merge Plan

No special instructions

Checklist

  • The PR has a short but descriptive title, suitable for a changelog
  • Tests added / updated (if applicable)
  • Documentation added / updated (if applicable)
  • Updated What's New copy (if doing a release after this PR)

@RyanJDick RyanJDick marked this pull request as ready for review January 28, 2025 22:36
@github-actions github-actions bot added python PRs that change python files backend PRs that change backend files labels Jan 28, 2025
@hipsterusername
Copy link
Member

Looks like this might not be kosher on mac, based on the tests

@RyanJDick
Copy link
Collaborator Author

Looks like this might not be kosher on mac, based on the tests

That's a known-flaky MPS test. It passed after re-running.

@hipsterusername hipsterusername force-pushed the ryan/tweak-model-cache-defaults branch from 93e219a to bf3ec57 Compare January 30, 2025 13:37
@hipsterusername hipsterusername enabled auto-merge (rebase) January 30, 2025 13:37
@hipsterusername hipsterusername merged commit f7315f0 into main Jan 30, 2025
15 checks passed
@hipsterusername hipsterusername deleted the ryan/tweak-model-cache-defaults branch January 30, 2025 13:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
backend PRs that change backend files python PRs that change python files
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants