A team is evaluating a RAG assistant. It wants to identify whether failures come from finding the wrong evidence or from writing unsupported answers despite receiving useful evidence. Which measurements should it include? Select 2 answers.

1. Only whether document ingestion completed without a service error.
2. Retrieval relevance or recall against known relevant passages for representative questions.
3. Only the number of words in each answer, with longer always considered better.
4. Only the storage size of the vector database.
5. Answer correctness and faithfulness to the supplied sources, alongside task-level outcomes.