Open SourceAnalytics16k
Langfuse
Open-source LLM engineering platform for tracing, evals, and prompt management.
Langfuse is an open-source observability and analytics platform purpose-built for LLM applications. It captures nested traces of agent and chain executions, tracks token usage and cost, and provides evaluation, prompt management, and dataset tooling. It integrates with the OpenAI SDK, LangChain, LlamaIndex, and OpenTelemetry, and can be self-hosted with Postgres and ClickHouse.
Repository
langfuse/langfuseLanguage
TypeScriptWhat you'd build with it
- Trace multi-step agent and RAG pipelines to debug latency, cost, and bad outputs
- Run offline and online LLM evals against curated datasets to catch regressions
- Version and A/B test prompts from a central UI without redeploying the app
Tags
llm-observabilitytracingevalsself-hosted