Langfuse
LLM observability with tracing, prompt management, evaluations and cost tracking.
Quick answer
Langfuse is an LLM observability with tracing, prompt management, evaluations and cost tracking, listed under LLM gateway & observability. A typical small-business use: see what your customer-facing bot actually answered and what it cost. Official hardware: recommended 4 vCPU · 16 GB RAM · 100 GB disk (no separate minimum published). That fits a Hetzner Cloud CPX42 server at about €69.99/month (server only, excl. VAT). License: MIT (core) + Langfuse Enterprise License (ee directories), which allows internal business use.
Last updated · facts verified against official docs on · sources
Activity data comes from GitHub and was last refreshed Oct 22, 2026.
What is Langfuse?
Langfuse records every LLM call your apps make (inputs, outputs, latency, tokens and cost) and adds prompt versioning, datasets and evaluations. It integrates with OpenAI SDKs, LangChain, LiteLLM and many frameworks.
Where it fits in a 10–200 person business
- See what your customer-facing bot actually answered and what it cost
- Version prompts and roll back bad changes
- Score answer quality before and after switching models
These use cases are our editorial suggestions, not claims made by the project.
Can a business use Langfuse? (license)
| License | MIT (core) + Langfuse Enterprise License (ee directories) |
|---|---|
| Type | Open core: Core is MIT/Apache; enterprise folders need a paid license. |
| Internal business use | Allowed under the license text |
| What to watch | Everything outside ee/, web/src/ee/ and worker/src/ee/ is MIT. Those directories need a commercial license. |
This is a plain-English summary of the license text, not legal advice. Third-party models you run with it have their own licenses.
What hardware does Langfuse need?
| Minimum | Not published |
|---|---|
| Recommended | 4 vCPU · 16 GB RAM · 100 GB disk |
For a single-VM Docker Compose deployment the docs recommend at least 4 cores and 16 GiB (e.g. AWS t3.xlarge) and ample storage, e.g. 100 GiB, because trace data grows quickly.
How much does it cost to host Langfuse per month?
| Recommended spec | 4 vCPU · 16 GB RAM · 100 GB disk → CPX42 (8 vCPU, 16 GB, 320 GB disk) €69.99/mo |
|---|
How we got this: we took the cheapest Hetzner Cloud Regular Performance (shared AMD vCPU) plan with at least the RAM and vCPU the project lists, using prices read from www.hetzner.com/cloud/regular-performance on 2026-10-22. "From" monthly price cap shown by Hetzner, excluding VAT. Prices vary by location. Covers the server only; it excludes LLM API usage, backups, extra storage, GPUs and your time. Disk is not matched; add a volume if you need more than the plan includes.
How to deploy Langfuse
Method: Docker Compose
git clone https://github.com/langfuse/langfuse.git
cd langfuse
# update the secrets in docker-compose.yml first
docker compose upFrom the official instructions: langfuse.com/self-hosting/deployment/docker-compose. Always check the source for the current version before you run it.
Known gotchas when self-hosting Langfuse
Alternatives in LLM gateway & observability
- LiteLLM AI gateway that puts 100+ LLM providers behind one OpenAI-compatible API with keys and budgets.
Langfuse FAQ
How much RAM does Langfuse need?
The recommended spec is 4 vCPU · 16 GB RAM · 100 GB disk. For a single-VM Docker Compose deployment the docs recommend at least 4 cores and 16 GiB (e.g. AWS t3.xlarge) and ample storage, e.g. 100 GiB, because trace data grows quickly.
How much does it cost to self-host Langfuse?
The software is free to download. Based on the official recommended 4 vCPU · 16 GB RAM · 100 GB disk (no separate minimum published): That fits a Hetzner Cloud CPX42 server at about €69.99/month (server only, excl. VAT). That covers the server only, not LLM API usage, backups, extra storage or your time.
Is Langfuse free for commercial use?
Langfuse is licensed under MIT (core) + Langfuse Enterprise License (ee directories) (Open core). Internal business use is allowed under the license text. Everything outside ee/, web/src/ee/ and worker/src/ee/ is MIT. Those directories need a commercial license. This is a summary, not legal advice.
What are common problems when self-hosting Langfuse?
Update the secrets in docker-compose.yml before the first start. There is no default login. The first user signs up in the UI. Docker's default json-file logging does not rotate logs, which can fill the disk unless you set max-size and max-file.
Is Langfuse still maintained?
Yes, it is actively developed. Last commit: Oct 22, 2026. Latest release: v4.67.0 (Oct 21, 2026). Checked Oct 22, 2026.
How do I install Langfuse?
The official method is Docker Compose. The exact commands are in the "How to deploy" section above and in the official instructions: https://langfuse.com/self-hosting/deployment/docker-compose
What are self-hosted alternatives to Langfuse?
Other apps in LLM gateway & observability: LiteLLM.
Sources
- GitHub repository: github.com/langfuse/langfuse
- Official documentation, Hardware requirements, Deploy instructions, Gotcha 1, Gotcha 2: langfuse.com/self-hosting/deployment/docker-compose
- License text: github.com/langfuse/langfuse/blob/main/LICENSE
- Hetzner Cloud pricing (cost estimate): www.hetzner.com/cloud/regular-performance
- GitHub stars, commits and releases: GitHub REST API, fetched
Facts on this page were checked against the project's own repository and docs on 2026-10-22. Spotted something out of date? Tell us.