Skip to content

1.2.5 - fix scheduler bug
Compare
Choose a tag to compare
@lukemarsden lukemarsden released this 15 Oct 11:59
· 17 commits to main since this release
1.2.5
6a937c0
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.

What's Changed

Fix bug that was causing requests to get matched to the wrong model instance.

  • fix(scheduler): reinstate model filter temporarily to ensure running models only get what they ask for by @philwinder in #533
  • Make context length smaller on 70b model so it fits in 48G by @lukemarsden in #534

Full Changelog: 1.2.4...1.2.5

Contributors

lukemarsden and philwinder
Assets 7
Loading