feat: Add multi-model support for LLM-as-a-judge evaluator #1196
+129
−11
Conversation
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
This change enables the LLM-as-a-judge evaluator to work with multiple model providers (OpenAI, Anthropic, Gemini, etc.) instead of only OpenAI. Key changes: - Switch to normalized API (uipath.llm.chat_completions) for multi-vendor support - Detect OpenAI models to conditionally apply response_format parameter - Add explicit JSON format instructions for non-OpenAI models in system prompt - Implement robust JSON parsing with markdown code block cleanup - Add comprehensive error and debug logging for troubleshooting This allows evaluators to use Anthropic Claude, Google Gemini, and other models configured in AllowedNormalizedModels. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Sonnet 4.5 <[email protected]>
github-actions
bot
added
test:uipath-langchain
- Fix bare except clause to catch Exception explicitly - Add type annotation for request_data dict to resolve mypy error - Apply ruff formatting fixes 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Sonnet 4.5 <[email protected]>
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Outdated suggestions cannot be applied.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.
Summary
This PR adds support for multiple model providers (OpenAI, Anthropic, Gemini, etc.) to the LLM-as-a-judge evaluator, enabling evaluations to work with any model configured in AllowedNormalizedModels.
Changes
uipath.llm.chat_completions(normalized API) for multi-vendor supportresponse_formatparameter (unsupported by Anthropic/Gemini)Testing Required
This PR requires testing with multiple model providers to ensure compatibility:
Test Files
Sample test configurations have been created in
samples/calculator/evaluations/:evaluators/llm-judge-semantic-similarity-claude.jsonevaluators/llm-judge-semantic-similarity-gemini.jsonevaluators/llm-judge-strict-json-similarity-claude.jsonevaluators/llm-judge-strict-json-similarity-gemini.jsoneval-sets/test-claude-evaluator.jsoneval-sets/test-gemini-evaluator.jsoneval-sets/test-gpt-evaluator.jsonBackward Compatibility
✅ Fully backward compatible with existing OpenAI-based evaluators. No breaking changes.
🤖 Generated with Claude Code