Open-source ML and LLM evaluation toolkit—drift detection, test suites, and dashboards for models in production.
Evidently AI is widely used for classical ML drift monitoring and has extended to LLM-specific checks (relevance, faithfulness, toxicity).