Evidently AI
Open-source framework for evaluating, testing, and monitoring the quality and safety of LLM applications and ML models in production.
About
Evidently AI is an open-source toolkit under the Apache 2.0 license that helps data scientists and MLOps engineers measure and track AI output quality throughout the full model lifecycle. It provides over 100 built-in metrics covering hallucination detection, PII identification, toxicity, retrieval quality, and data drift, plus tools for generating synthetic test cases and adversarial inputs. Live dashboards let teams monitor evaluation metrics in production and catch degradation before it affects users. The core framework is free, and the company supports itself through community engagement and enterprise offerings for teams that need managed monitoring infrastructure.