AI · Early access

Langfuse hosting on your own server

An open-source observability and tracing platform for LLM applications.

Early access. Join the list and we email you when Langfuse opens on VPS.

What Langfuse does

Langfuse tracks prompts, model calls, latency, and cost across an LLM application, giving developers a trace view similar to APM tools but tuned for AI workloads. Teams use it to debug agent behavior, run evaluations, and monitor production LLM apps. It also supports prompt versioning and dataset-based evaluation, so teams can compare prompt changes against real production traffic.

Langfuse at a glance

License: MIT (core), proprietary Enterprise features
Source code: github.com/langfuse/langfuse
Website: langfuse.com
Runs on: Your own server (VPS)
Good to know: Needs Postgres and ClickHouse; trace volume drives storage and RAM needs more than compute.

Debugging LLM app behavior
Prompt and cost monitoring
Running LLM evaluation pipelines

Langfuse, connected to the rest of your business

On your domain
Langfuse answers at an address like langfuse.yourbusiness.com, with SSL and daily backups switched on.
Mail from your address
Langfuse sends its emails from your CloudWish business mailbox. Business email
AI with a cap you set
Give Langfuse its own AI key with a monthly spend cap. Usage shows as its own line on your bill. AI in your apps (early access)