promptfoo
Turn real AI-agent failures into reviewed regression tests, and measure whether later versions fix or reintroduce them.
Statistics for LLM-judged evaluations: bias-corrected scores, real confidence intervals, and comparisons that hold up
Translate one rubric-spec rubric and run configuration into supported evaluation-framework inputs and exports.
Prompt optimization and benchmarking for local LLMs: picks the right prompt technique for your task and measures it on Ollama. DSPy-compatible.
