feat: Add --judge CLI argument and require full provider prefix for model IDs#29
Open
yudistiraashadi wants to merge 2 commits intopinchbench:mainfrom
Open
Conversation
Introduced a new command-line argument `--judge` to specify the judge model identifier, defaulting to "openrouter/anthropic/claude-opus-4.5". Updated grading logic to include the judge model if provided, enhancing flexibility in task evaluation.
Changed model identifiers in benchmark commands to include the 'openrouter/' prefix. Added a note about the requirement for provider prefixes in model IDs. Enhanced the table formatting for better readability.
Contributor
Code Review SummaryStatus: No Issues Found | Recommendation: Merge Other Observations (not in diff)Minor inconsistencies found in unchanged code that cannot receive inline comments:
These are documentation-only nits and do not affect functionality. Files Reviewed (4 files)
|
Member
|
@yudistiraashadi thanks for this! How would you compare this to #27 ? |
Author
I think #27 includes changes in this PR + more. If you merge #27 , we can safely close this PR. |
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Motivation
Currently, model IDs are normalized internally which limits running benchmarks to a specific provider. By requiring full provider prefixes and adding the
--judgeflag, users can run benchmarks from any provider they want. For example, using Qwen models from a different provider, or any other model/provider combination — giving full flexibility over both the benchmarked model and the judge model.Summary
--judgeCLI argument tobenchmark.pyto allow customizing the judge model used for gradingopenrouter/anthropic/claude-3.5-sonnet) instead of normalizing them internally — removesnormalize_model_id()fromlib_agent.pylib_grading.pyto useopenrouter/prefix--judgeflag documentation, and formatting improvements