Langfuse

license

MIT(core:tracing/integrations/API/data-model/exports/evals/playground/SSO)

commercial

enterprise-security(SCIM,audit-logs,data-retention)

source

public

core features withheld

yes — the root LICENSE carves ee/, web/src/ee/ and worker/src/ee/ out of MIT and under ee/LICENSE, the Langfuse Enterprise License

Langfuse's core is MIT-licensed and runs without limits or a license key, but the ee/, web/src/ee/ and worker/src/ee/ directories sit under a separate Enterprise License. Features such as audit logs, data retention, SCIM, the admin API and UI customization need a license key when self-hosting, which is why Langfuse is open core rather than fully open source.

• https://langfuse.com/self-hosting/license-key recorded 2026-09-29

  Self-hosting license-key page. It says "All core Langfuse features and APIs are available in Langfuse OSS (MIT licensed) without any limits" and lists the features that need a key, including "Audit Logs" and "Data Retention Policies", and adds "All core OSS features continue to work without a license key".

• https://raw.githubusercontent.com/langfuse/langfuse/main/ee/LICENSE recorded 2026-09-29

  The Langfuse Enterprise License, read in full. It says "Certain parts of the periphery of Langfuse are commercially licensed and governed by this Enterprise License" and that the software "may only be used, if you (and any entity that you represent) have agreed to, and are in compliance with, the applicable agreement governing the use of the Software".

• https://raw.githubusercontent.com/langfuse/langfuse/main/LICENSE recorded 2026-09-29

  Root LICENSE, read in full. It says "Portions of this software are licensed as follows:" and "Content outside of the above mentioned directories or restrictions above is available under the" MIT Expat license, while the ee/, web/src/ee/ and worker/src/ee/ directories fall under ee/LICENSE.

• https://raw.githubusercontent.com/langfuse/langfuse/main/README.md recorded 2026-09-29

  Repository README: "Langfuse can be **self-hosted in minutes** and is **battle-tested**" and "Run Langfuse on your own machine in 5 minutes using Docker Compose.", so the whole platform is published and self-hostable.

The langfuse PyPI package is a client: applications install it to send traces, so its downloads count instrumented applications rather than Langfuse deployments. ClickHouse's acquisition announcement is the evidence that speaks to the product, reporting 6M+ pulls of the server image, with no period stated, and use by 19 of the Fortune 50. Those are the vendor's claims rather than counts made here, and a pull is not a deployment.

• https://clickhouse.com/blog/clickhouse-acquires-langfuse-open-source-llm-observability recorded 2026-09-21

  'ClickHouse has acquired Langfuse, the leading open-source platform for LLM observability, evaluations, and prompt management', with '23.1M+ SDK installs per month', '6M+ Docker pulls', '20k+ GitHub stars' and 'Trusted by 19 of the Fortune 50 and 63 of the Fortune 500'. The Docker figure is the one that counts the server rather than the client.

• https://github.com/langfuse/langfuse recorded 2026-08-15

  Repository page for langfuse/langfuse: 33.1k stars, 3.6k forks, MIT licence, public and unarchived. Corroboration only — a star count is not a use and bands nothing.

• https://pypistats.org/api/packages/langfuse/recent recorded 2026-08-14

  last_month = 26,611,793 downloads for langfuse (last_week 6,315,194, last_day 1,118,108).

Langfuse combines tracing, prompt management, evaluation, datasets and a prompt playground in one self-hostable platform. There is no external benchmark for LLM observability, so most other products in this category are compared directly against this feature set.

• https://github.com/langfuse/langfuse recorded 2026-08-13

  Repository overview. The README still describes tracing, prompt management, evaluations, datasets and a playground as one platform, with OpenTelemetry and framework integrations.

• https://langfuse.com/docs recorded 2026-09-29

  Docs home: "Evaluate prompts and models on datasets directly in the user interface. No custom code is needed."

• https://langfuse.com/docs/prompt-management/overview recorded 2026-09-29

  Prompt management overview: "Prompt management is a systematic approach to storing, versioning, and retrieving prompts for your LLM application."