Expand description
Evaluator for LLM providers Module for evaluating and comparing responses from multiple LLM providers.
This module provides functionality to run the same prompt through multiple LLMs and score their responses using custom evaluation functions.
Structs§
- Eval
Result - Result of evaluating an LLM response
- LLMEvaluator
- Evaluator for comparing responses from multiple LLM providers
- Parallel
Eval Result - Result of a parallel evaluation including response, score, and timing information
- Parallel
Evaluator - Evaluator for running multiple LLM providers in parallel and selecting the best response
Type Aliases§
- Scoring
Fn - Type alias for scoring functions that evaluate LLM responses