By arize-ai

arize-evaluator

Build and run evaluators across traces and experiments.

LLM evaluation

Only the most recent Revisions are kept. Restoring one saves its content again as the newest.