1.2.10 - slimmer small image, fix scheduler regression

What's Changed

Fix issue where scheduler doesn't progress when there are no new requests from clients. Also, slim down the small image to make waiting less tedious on initial install.

• revert(scheduler): do re-scheduling in the runner poll by @philwinder in #538

Full Changelog: 1.2.9...1.2.10

1.2.9 - faster docker images & new models

What's Changed

New base layers for faster model pulling in small image. Add mistral-nemo and q6 quant of mixtral to large image.

Full Changelog: 1.2.8...1.2.9

1.2.8 - fix eviction in runners

What's Changed

Fixes a bug that was causing eviction to fail and thus scheduling to get clogged up and hang.

• make eviction more reliable by @lukemarsden in #537

Full Changelog: 1.2.7...1.2.8

1.2.7 - fix result limits in knowledge/RAG

What's Changed

• fix/rag result count limit by @rusenask in #536

Full Changelog: 1.2.6...1.2.7

1.2.6 - return scheduling errors in the UI

What's Changed

• fix(errors): report errors on streaming responses by @philwinder in #535

Full Changelog: 1.2.5...1.2.6

1.2.5 - fix scheduler bug

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

1.2.4 - include llama3:instruct in base image

What's Changed

• include llama3:instruct in base image by @lukemarsden in #532

Full Changelog: 1.2.3...1.2.4

1.2.3 - fix version information

What's Changed

Version information is now correctly shown in the dashboard.

• tag versions in the right way by @lukemarsden in #529
• put this in the right place by @lukemarsden in #530

Full Changelog: 1.2.2...1.2.3

1.2.3-rc2 - another version fix

What's Changed

• put this in the right place by @lukemarsden in #530

Full Changelog: 1.2.3-rc1...1.2.3-rc2

1.2.3-rc1 - test version tagging

What's Changed

• tag versions in the right way by @lukemarsden in #529

Full Changelog: 1.2.2...1.2.3-rc1