
Langfuse
β Text source-checked Nov 3, 2026Open-source AI engineering platform for building, monitoring, and improving LLM applications.
Source-checked against the vendor's site Β· rebuilt from the vendor's own pagesProfile updated Link checked Pricing captured How we rate
In short
Langfuse is an open-source platform for building and monitoring LLM apps. It offers tracing, prompt management, and evaluation with flexible self-hosting options.
Langfuse is an open-source AI engineering platform that helps teams build, monitor, and improve their LLM applications.
The vendor says that hierarchical traces capture every LLM call, tool invocation, and retrieval step. Prompt management allows users to separate prompts from code with one-click deployments and rollbacks. Evaluation features include LLM-as-a-judge, heuristic functions, or human review.
Langfuse works with any language and framework supporting OTel instrumentation. The platform offers over 100 integrations to facilitate setup.
Self-hosting is available via Docker Compose, Kubernetes, and Terraform templates for AWS, GCP, and Azure.
The Hobby plan is free and does not require a credit card. The vendor says that the Core plan costs $29 per month and includes longer data access and unlimited users.
According to Langfuse: Langfuse is an open-source AI engineering platform that helps teams build, Monitor, and improve their LLM applications. Its pages list this strength: The codebase is MIT licensed.
β The AI Tool Bible editorial team
Pros
- β The codebase is MIT licensed.
- β Langfuse can be self-hosted using Docker on user infrastructure.
- β The platform supports any language and framework with OTel instrumentation.
- β Over 100 integrations are available for setup.
- β A free Hobby plan is available without a credit card.
Pricing (captured )
| Tier | Monthly | Key features |
|---|---|---|
| Hobby | Free |
|
Prices change; we show the grid as captured on Oct 14, 2026. Check current pricing on the vendor's site β
Facts
- Platforms
- api (not source-checked)
- API available
- Yes (not source-checked)
- Free trial
- Yes (not source-checked)
- Open source
- Yes (license: NOASSERTION)
- GitHub stars
- 37,371 (checked Nov 13, 2026) (not source-checked)
- Last GitHub push
- Nov 12, 2026 (not source-checked)
- First commit
- May 2023 (not source-checked)
- Model used
- Model-agnostic (not source-checked)
- Docs pages indexed
- 105 pages we found in the docs (not source-checked)
- Package downloads (30 d)
- 31,213,932 Β· pypi langfuse Β· checked Nov 14, 2026 (not source-checked)
- Hacker News
- 28 stories Β· top 220 points Β· last Sep 17, 2026 Β· checked Nov 14, 2026
Where this tool sits
Use cases
Frequently asked
- How much does Langfuse cost?
- Langfuse offers a free Hobby plan. Paid plans include Core at $29/month, Pro at $199/month, and Enterprise at $2,499/month. The Hobby plan does not require a credit card.
- Can I self-host Langfuse?
- Yes, self-hosting is available via Docker Compose, Kubernetes, and Terraform templates. The platform supports deployment on AWS, GCP, and Azure environments for full control.
- Does Langfuse work with my existing framework?
- Langfuse is model-agnostic and works with any language or framework supporting OTel instrumentation. The platform offers over 100 integrations to facilitate easy setup and connection.
- What evaluation methods does Langfuse support?
- Evaluation features include LLM-as-a-judge, heuristic functions, or human review. These tools help teams monitor and improve LLM application performance effectively.
- How does prompt management work in Langfuse?
- Prompt management allows users to separate prompts from code. It supports one-click deployments and rollbacks, facilitating efficient prompt testing and model comparison workflows.