AI Eval

AI observability & eval

Shipping an LLM product? Tracing, prompt versioning, scorers, and regression eval tools.

12 entries

Observability

7 entries

Open-source LLM tracing, prompt mgmt, evals. Self-hostable.

Drop-in proxy for OpenAI/Anthropic/etc; logs, costs, caching.

Prompt versioning + observability + A/B testing.

LangChain's own platform for tracing, datasets, evals, monitoring.

Open-source observability for LLM, embedding and agent traces.

Open-source LLM analytics, prompt mgmt, user-level tracing.

Conversation observability + evals + dataset curation.

5 entries

End-to-end evals for AI products: prompts, datasets, scorers, regressions.

OpenAI's reference framework for benchmark + scenario evals.

Local-first eval runner; YAML-defined test suites against any provider.

Pytest-style metrics for hallucination, faithfulness, relevancy.

Reference framework for RAG-pipeline evaluation metrics.