Langfuse

An LLM engineering platform for tracing, prompt management and evaluation of AI applications, which you can self-host or use in the cloud.

36.8k stars, 4.1k forks TypeScript On GitHub since May 2023

Visit website View on GitHub How to run it

What is Langfuse?

Langfuse records traces of LLM applications, including model calls, retrieval, embeddings and agent steps, so you can inspect sessions and see where an answer went wrong. Prompt management versions prompts centrally with caching, and evaluations combine LLM-as-a-judge, code checks, user feedback and manual labels. Datasets hold test sets for experiments before a release, and a playground lets you try prompts and model settings straight from a bad trace. SDKs for Python and JavaScript, an OpenAPI spec and integrations with frameworks such as LangChain and LlamaIndex connect it to your code.

Good fit if

  • You build chatbots, RAG systems or agents and need to see what each request did and what it cost.
  • Your team wants to version prompts and test changes before they reach users.

Look elsewhere if

  • You need error tracking for ordinary application code: Sentry.
  • You want product analytics and session replay with some LLM metrics in the same tool: PostHog.

Before you adopt it

Langfuse is MIT-licensed except for the ee folders, which carry a separate commercial license, so some features are paid. Outside contributions require signing a CLA. Since January 2026 the project has been part of ClickHouse. The Docker Compose setup runs six services, including ClickHouse, PostgreSQL, Redis and MinIO.

Drafted with AI from the project's README and repository data, then checked against that data on 31 October 2026. Report a correction

Langfuse health: is it still maintained?

80 of 100, actively developed

  • Activity
    2,266 commits in the last 90 days, the latest on 1 November 2026.
    30 / 30
  • Contributors
    15 people wrote the latest 100 commits; the most active wrote 24% of them.
    16 / 20
  • Releases
    15 releases in the last 12 months; the latest, v4.72.0, on 29 October 2026.
    19 / 20
  • Replies to issues
    Too few recent issues from outside the team to measure, so this part gets a neutral score.
    10 / 20
  • Security practices
    No OpenSSF Scorecard for this repository yet, so this part gets a neutral score.
    5 / 10

Read from GitHub on 1 November 2026. How the score works

Langfuse license: is it open source?

Langfuse uses MIT, an open-source license approved by the OSI. Some features sit under a separate paid license inside the same repository. Contributors sign a contributor license agreement, which lets the owner relicense their work later.

  1. May 2023First published on GitHub.
  2. TodayMIT, with paid parts: license file assigns a different licence to part of the code; directory ee/; directory packages/shared/src/server/ee/; directory web/src/ee/; directory worker/src/ee/. A contributor license agreement is required.

How to run Langfuse yourself

Docker
Dockerfile or compose file in the repository
Compose file
docker-compose.yml: 6 services (clickhouse, minio, redis, postgres)
Setup effort
5 of 5, judged from the compose file
Kubernetes
No Helm chart in the repository
Desktop and mobile
No app builds in GitHub releases
Interface languages
English only

Install instructions: langfuse.com or the README on GitHub.

Alternatives to Langfuse

  • SentryDeveloper-first error tracking and performance monitoring 85 health · Functional Source License · 45.7k stars
  • code-serverVS Code in the browser 85 health · MIT · 79.7k stars

All 3 projects in Developer tools

Questions about Langfuse

Is Langfuse open source?

Yes. Langfuse is published under MIT, an OSI-approved open-source license. Some features are kept under a separate paid license in the same repository.

Is Langfuse still maintained?

Its health score is 80 of 100 (actively developed). It had 2,266 commits in the last 90 days, the latest on 1 November 2026, and its latest release came out on 29 October 2026.

How do I run Langfuse?

You can run it with Docker. The README on GitHub has the exact steps.

What are the alternatives to Langfuse?

Close alternatives in the Developer tools category include Sentry, code-server.