Secure & reliable LLMs | promptfoo

Eliminate risk with AI red-teaming and evals used by 80,000+ developers. Find and fix vulnerabilities, maximize output quality, catch regressions.