Evals & observability
Langfuse
Open-source, self-hostable tracing, prompt management, and evaluation for model applications.
- Category
- Evals & observability
- Pricing
- Open source
- Runs
- Hosted or self-hosted
- Languages
- Python, TypeScript
- Interface
- SDK, Dashboard
- Source
- Open source
What Langfuse is
Langfuse covers tracing, prompt versioning, evaluation, and cost tracking, and is genuinely self-hostable rather than open-source-in-name. Prompt management is the part teams underrate: prompts live outside your deploy cycle, versioned, so changing one is not a release.
Best for
Teams who need traces containing customer data to stay inside their own infrastructure.
Consider something else if
Self-hosting means running a database and a service you now own.
Langfuse alternatives
The closest options in evals & observability, on the axes that actually separate them.
| Tool | Best for | Pricing | Runs |
|---|---|---|---|
| Langfuseopen source | Teams who need traces containing customer data to stay inside their own infrastructure. | Open source | Hosted or self-hosted |
| LangSmith | LangChain and LangGraph users, who get the deepest integration for the least work. | Free tier | Hosted |
| Phoenixopen source | Teams standardised on OpenTelemetry, and anyone debugging retrieval quality interactively. | Open source | Hosted or self-hosted |
| Heliconeopen source | Getting cost and latency visibility across a codebase quickly, especially one you do not want to instrument. | Free tier | Hosted or self-hosted |
| Braintrust | Teams treating prompt changes as changes that need tests, with evals wired into CI. | Free tier | Hosted |
| W&B Weaveopen source | Organisations already on Weights & Biases who want one place for all model work. | Free tier | Hosted |
Choosing within evals & observability
Tracing or evaluation first
Some tools lead with tracing and add evals; others are eval-first with tracing attached. If you are debugging today, start with tracing. If you are about to change a prompt that is already in production, start with evals.
How it gets your data
A proxy needs a one-line base-URL change and captures everything. An SDK needs instrumentation and captures more structure. OpenTelemetry-based tools slot into tracing you may already run.
Questions
What is Langfuse?
Langfuse covers tracing, prompt versioning, evaluation, and cost tracking, and is genuinely self-hostable rather than open-source-in-name. Prompt management is the part teams underrate: prompts live outside your deploy cycle, versioned, so changing one is not a release. It is open source and hosted or self-hosted.
What are the alternatives to Langfuse?
The closest alternatives are LangSmith, Phoenix, Helicone, Braintrust, W&B Weave. They sit in the same category — evals & observability — and differ mainly on hosting model, pricing shape, and how much they abstract away.
Is Langfuse the right choice?
Teams who need traces containing customer data to stay inside their own infrastructure. The main caveat: Self-hosting means running a database and a service you now own.
Whatever you build on, the model is the line item that scales. See what each one costs per million tokens, or price your own workload.