Langfuse
Open-source LLM engineering platform for tracing, evaluation, and prompt management.
Langfuse is profiled here as a Observability tool for engineering teams. Read about features, pricing, and how it compares to related options in the tools directory.
Description
Langfuse is an open-source LLM observability and engineering platform founded in 2023 by Marc Klingen, Maximilian Deichmann, and Clemens Rawert through Y Combinator's W23 batch. ClickHouse acquired the company in January 2026 and continues to operate it as a standalone product. The platform gives engineering teams the tooling to debug agent flows, version prompts, and evaluate LLM outputs across providers including OpenAI, Anthropic, Cohere, and self-hosted models.
LLM tracing with hierarchical spans for nested agent and retrieval steps
Prompt versioning and prompt management with low-latency caching
LLM evaluation through LLM-as-a-judge, user feedback, and manual labeling
Datasets for offline testing and regression checks across application versions
OpenTelemetry-native SDK with Python and TypeScript support
Integrations with LangChain, LlamaIndex, Haystack, OpenAI SDK, and LiteLLM
Self-hostable via Docker Compose, single VM, or Kubernetes Helm chart
Open source under MIT license with a free cloud tier
Alternative tools
- HoneyHive
Evaluation and observability platform for AI agents
- Sentry
Error tracking and performance monitoring for developers
- SigNoz
Open-source, OpenTelemetry-native observability platform
- Datadog
Unified observability for metrics, traces, and logs
- Arize AX
Enterprise platform for AI observability and evaluation
- OpenTelemetry
Vendor-neutral standard for traces, metrics, and logs