Comet Opik

Opik (by Comet): The open-source, end-to-end platform for rigorous LLM evaluation, tracing, and production monitoring.

This is Opik: an open-source, developer-first platform from Comet, engineered for high-stakes LLM and RAG application workflows. It delivers comprehensive observability: tracing every LLM call, logging conversations, and monitoring agent activity across the entire pipeline. Developers use its robust evaluation framework, including LLM-as-a-judge metrics and Pytest integration, to automate testing and confidently catch critical issues like hallucinations before deployment. Opik ensures your GenAI applications are reliable, performing optimally, and production-ready.

https://www.comet.com/site/products/opik/