Repository navigation
Use Gemini 3.8 Flash as the default audio judge - #263
Open
fanny-riols wants to merge 11 commits into
Open
fanny-riols wants to merge 11 commits into
fanny-riols wants to merge 11 commits into
Conversation
fanny-riols
changed the base branch from
main
to
pr/fr/judge-language-guidance-content
November 12, 2026 15:02
fanny-riols
marked this pull request as ready for review
November 12, 2026 16:45
fanny-riols
force-pushed
the
pr/fr/audio-judge-gemini-3-8-flash
branch
from
November 12, 2026 16:49
d1cf35a to
708d485
Compare
| type: 'lalm_judge', | ||
| judgeModel: 'Gemini 3 Flash', | ||
| judgeModel: 'Gemini 3.8 Flash', | ||
| judgeAccuracy: 0.8957, |
Collaborator
There was a problem hiding this comment.
do the accuracy labels need updating?
Collaborator
Author
There was a problem hiding this comment.
Yep, it's one of the PR description todos :)
…ntent' into pr/fr/audio-judge-gemini-3-8-flash # Conflicts: # tests/fixtures/metric_signatures.json
…ntent' into pr/fr/audio-judge-gemini-3-8-flash
…ntent' into pr/fr/audio-judge-gemini-3-8-flash # Conflicts: # tests/fixtures/metric_signatures.json
katstankiewicz
approved these changes
Nov 12, 2026
katstankiewicz
left a comment
katstankiewicz
left a comment
Collaborator
There was a problem hiding this comment.
approved when judge accuracy is updated and website is re-built 😄
…ntent' into pr/fr/audio-judge-gemini-3-8-flash # Conflicts: # tests/fixtures/metric_signatures.json
…ntent' into pr/fr/audio-judge-gemini-3-8-flash # Conflicts: # tests/fixtures/metric_signatures.json
…ntent' into pr/fr/audio-judge-gemini-3-8-flash
…ntent' into pr/fr/audio-judge-gemini-3-8-flash
…ntent' into pr/fr/audio-judge-gemini-3-8-flash # Conflicts: # tests/fixtures/metric_signatures.json
…ntent' into pr/fr/audio-judge-gemini-3-8-flash
…ntent' into pr/fr/audio-judge-gemini-3-8-flash
This branch has not been deployed
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
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.
Why
Gemini 3.8 Flash is a better judge than Gemini 3.0 Flash.
What changes
AudioJudgeMetric.default_model:gemini-3-flash-preview→gemini-3.8-flash, with the same parameters.docs/metrics/README.md, the three speech-fidelity metric pages and the judge table indocs/experiment_setup.md.website/src/data/metricsData.ts). The site is not rebuilt here.tests/fixtures/metric_signatures.jsonregenerated.Before merging