Opik¶

Opik serves as a comprehensive platform for evaluating LLMs. It enables you to confidently assess, test, and deploy LLM applications, equipped with a range of observability tools designed to fine-tune language model outputs throughout both development and production phases.

Opik can be used for:

• Observability: Log all your LLM calls and chains during development and in production
• Evaluation: Store your evaluation datasets in Opik and easily evaluate the performance of your LLM applications using Opik's built-in evaluation metrics (Hallucination, Context Relevance, and more) or using custom metrics
• Testing: Use Opik's integration with PyTest to automate the testing of your LLM application before it is deployed to production
• Production: Monitor and debug your LLM applications in production

Getting started¶

Explore the following guides to get started with Opik:

• Getting started with Opik
• Logging LLM calls
• Opik's integrations (LangChain, OpenAI, LlamaIndex and more)
• Evaluating your LLM applications