feat(evaluate): support llm rater configuration in evalbench generator#64
Merged
wangauone merged 5 commits intofeat/hill-climbingfrom Apr 3, 2026
Merged
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request introduces an LLM-based rater configuration to the evaluation generator, adding a new llmrater_config.yaml file and updating the run configuration to include set_match and llmrater scorers. Feedback suggests retaining the previously removed deterministic scorers (exact_match and executable_sql) to maintain comprehensive evaluation, avoiding the use of hardcoded preview model versions in the generator, and enhancing tests to verify the actual content of the newly generated configuration.
helloeve
approved these changes
Apr 3, 2026
wangauone
added a commit
that referenced
this pull request
Apr 21, 2026
🤖 I have created a release *beep* *boop* --- ## [0.5.0](v0.4.3...v0.5.0) (2026-04-21) ### Features * **autoctx:** introduce automated context generation ([#95](#95)) ([57f8aa7](57f8aa7)) * **bootstrap:** enrich workflow with user-provided docs and code ([#70](#70)) ([04c8455](04c8455)) * **bootstrap:** enrich workflow with user-provided docs and code ([#72](#72)) ([bd04743](bd04743)) * **evaluate:** add custom runner configs to lower evalbench parallelism ([#65](#65)) ([f0758fa](f0758fa)) * **evaluate:** populate gcp_project_id in llmrater config ([#90](#90)) ([016af70](016af70)) * **evaluate:** simplify golden dataset format and write configs dire… ([#68](#68)) ([a76b4bb](a76b4bb)) * **evaluate:** support llm rater configuration in evalbench generator ([#64](#64)) ([d181b82](d181b82)) * **evaluate:** unify failure reporting and add execution errors ([#80](#80)) ([c11ff47](c11ff47)) * **evaluate:** use gemini-2.5-pro for evaluation rater ([#67](#67)) ([5c8f92d](5c8f92d)) * **facet:** enforce qualified table names in prompts ([#85](#85)) ([b37480f](b37480f)) * **hillclimb:** support eval result reading tool ([#69](#69)) ([bb5a249](bb5a249)) * **mcp:** add autoctx-hillclimb workflow skill and tools ([#59](#59)) ([2236964](2236964)) * **mcp:** adopt ADC support for Cloud SQL and AlloyDB ([#78](#78)) ([f4091fa](f4091fa)) * **mcp:** centralize Gemini model configuration ([#73](#73)) ([97c5c25](97c5c25)) * **mcp:** migrate autoctx infrastructure to autoctx/ folder ([#92](#92)) ([0b4f52d](0b4f52d)) * **mcp:** switch default model to gemini-2.5-flash ([#93](#93)) ([60800e5](60800e5)) * **mcp:** update evalbench version to 1.4.0 ([#83](#83)) ([0b4ffb3](0b4ffb3)) * **skills:** update bootstrap skill to include upload URL ([#77](#77)) ([fd65bff](fd65bff)) ### Miscellaneous Chores * force release version 0.5.0 ([da04e7c](da04e7c)) * force release version 0.5.0 ([#97](#97)) ([82364d6](82364d6)) --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please).
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.
Adds support for the EvalBench LLM Rater to enable semantic result comparisons during automated evaluations.
llmrater_config.yamlusinggemini-3.1-pro-preview.run_config.yamlscorers definition.