LLM-as-a-Judge and RAG Evaluations on Amazon Bedrock

Bedrock RAG Evaluations

LLM-as-a-judge and End-to-End RAG evaluations are now generally available on Amazon Bedrock.

  • Evaluate FMs to select the best one for your use case
  • Compare results across multiple evaluation jobs to make decisions faster
  • Evaluate your end-to-end RAG workflow
  • Ensure complete and relevant retrieval from your RAG system