Default the CI model to Qwen, the model this build is measured on - #70
Merged
Merged
Conversation
The task-battery and a11y-audit workflows fell back to Llama-3.3-70B when no ELM_MODEL secret was set, and the README's .env.local example named it too — both leftovers from before v4.0.0 moved the app to Qwen. With the repository secret about to carry only the gateway URL and key, the default is what CI runs, and it has to be the model production runs: the startup report in every recent battery job said "Model resolution is off the measured profile", and the L2/L3/L5 language tasks fail on the fallback model while passing on Qwen.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The task-battery and a11y-audit workflows fell back to
meta-llama/Llama-3.3-70B-Instructwhen noELM_MODELsecret was set, and the README's.env.localexample named it too — leftovers from before v4.0.0 moved the app to Qwen. With the repository secrets about to carry only the gateway URL and key, the workflow default is what CI runs, and it has to be the model production runs.Every recent battery job's startup report said "Model resolution is off the measured profile", and the language tasks from #68 (L2/L3/L5/L6) fail intermittently on the fallback model while passing 6/6 on Qwen — which is why
mainhas been red since v4.2.19 with T1–T7 all green.Three files, five lines: the two workflow fallbacks and the README example. Not included: making the off-profile warning fail the battery job outright — worth doing separately if we want CI to refuse to grade the wrong model.