Langfuse review

Q: How do I install Langfuse?

Run pip install langfuse openai. It needs Node.js >=18. The install section above has the other options from the project's README.

Langfuse is an open source LLM engineering platform for tracing, evaluating, prompt management, and debugging LLM applications and agents.

Updated Nov 1, 2026. Review written Oct 22, 2026 from the project's own sources; usage data refreshed every Monday.

TL;DR

Langfuse is an open source platform for tracing, evaluating, and improving LLM applications and AI agents. Langfuse is for teams building LLM apps who want observability, prompt management, evaluations, and datasets in one place. Langfuse can run as a managed cloud service or be self-hosted with Docker Compose, Kubernetes, or Terraform.

36.8kGitHub stars
+414new stars, week of Oct 25
~9.8Mdownloads a week, +9% vs 4-week avg
#4in Evals and observability this week

Langfuse at a glance

Category: Evals and observability
Maintained by: Langfuse (organization, @langfuse on GitHub)
Language: TypeScript, Rust, JavaScript
Repository: github.com/langfuse/langfuse
Website: langfuse.com
npm package: langfuse 3.39.2
PyPI package: langfuse 5.0.4
Latest release: v4.67.0 on GitHub, Oct 21, 2026
First published: May 18, 2023

What Langfuse does

Langfuse ingests traces of LLM calls, retrieval, embedding, and agent actions so you can inspect and debug logs and user sessions.
Langfuse manages prompts centrally with version control, using server- and client-side caching to avoid adding latency.
Langfuse supports LLM-as-a-judge, code evaluators, user feedback, manual labeling, and custom evaluation pipelines via APIs and SDKs.
Langfuse datasets provide test sets and benchmarks for pre-deployment testing and structured experiments.
Langfuse includes an LLM Playground for testing prompts and model configurations, reachable directly from a bad trace result.
Langfuse offers a public API with an OpenAPI spec, a Postman collection, and typed SDKs for Python and JS/TS.

How it works

Langfuse runs as Langfuse Cloud or on your own infrastructure. You instrument your app with the Python or JS/TS SDKs, the @observe() decorator, or integrations such as OpenAI, LangChain, LlamaIndex, Haystack, LiteLLM, Vercel AI SDK, and Mastra. Traces go to the Langfuse server, which is built with ClickHouse. API keys and a base URL connect your app.

How to install and start using Langfuse

Python SDK with OpenAI integration

pip install langfuse openai

JS/TS SDK (current)

npm install @langfuse/tracing @langfuse/otel @opentelemetry/sdk-node

Self-host locally with Docker Compose

docker compose up

Requirements

The Python SDK requires Python 3.11 or later and below 4.0.
The npm langfuse package requires Node.js 18 or later.
Langfuse API credentials (secret and public keys) from a project.
Docker Compose for local or VM self-hosting, or Kubernetes with Helm for production.

First steps

Create a Langfuse Cloud account or self-host Langfuse, then create a new project.
Create new API credentials in the project settings.
Set LANGFUSE_SECRET_KEY, LANGFUSE_PUBLIC_KEY, and LANGFUSE_BASE_URL in a .env file, choosing the EU or US region.
Wrap functions with the @observe() decorator and use the Langfuse OpenAI integration, then view traces in Langfuse.

Who Langfuse is for

Teams debugging LLM applications and agents in production through hierarchical traces.
Developers who want to self-host LLM observability on their own infrastructure.
Teams versioning prompts and running evaluations and experiments on datasets.
Projects built on LangChain, LlamaIndex, OpenAI SDK, or Vercel AI SDK that want automated instrumentation.

Limitations to know

Langfuse is MIT licensed except for the ee folders, which fall under separate terms.
The npm package langfuse is a deprecated v3 client; new JS/TS work should use @langfuse/tracing and @langfuse/otel.
Self-hosting with the default docker-compose.yml can exhaust disk space unless Docker log rotation is configured.
The LlamaIndex and Haystack integrations support Python only; the LiteLLM JS/TS integration is proxy only.

Is Langfuse growing? Usage data

In the week of Oct 25–31, 2026, Langfuse ranked #71 on the Devtoolwatch Board and #4 in Evals and observability, with a momentum score of 70 out of 100. Its main package had about 9.8M downloads that week, +9% against its own 4-week average.

New GitHub stars per week for the last 12 weeks, from GitHub's stargazer history (today's stargazers by the week they starred).

Show the weekly numbers

New GitHub stars per week for Langfuse
Week of	New stars
Aug 9, 2026	415
Aug 16, 2026	421
Aug 23, 2026	362
Aug 30, 2026	346
Sep 6, 2026	292
Sep 13, 2026	314
Sep 20, 2026	274
Sep 27, 2026	276
Oct 4, 2026	299
Oct 11, 2026	377
Oct 18, 2026	444
Oct 25, 2026	414

On Hacker News

ClickHouse acquires Langfuse, 220 points and 96 comments (Jan 17, 2026)
Launch HN: Langfuse (YC W23) – OSS Tracing and Workflows to Improve LLM Apps, 215 points and 61 comments (Dec 17, 2024)
Show HN: Langfuse – Open-source observability and analytics for LLM apps, 143 points and 35 comments (Aug 29, 2023)
Doubling Down on Open Source, 137 points and 13 comments (Jun 4, 2025)
Building (Open Source) Custom Dashboards Is Harder Than You Think, 33 points and 0 comments (May 21, 2025)

The Board ranks tools by four signals each week: new stars, new stars for the repo's size, added downloads and download growth beyond the market. How the ranking works.

Langfuse alternatives

Other LLM evals and observability tools on the Board in the week of Oct 25–31, 2026, by momentum:

Evals and observability ranked by momentum, the week of Oct 25–31, 2026
Rank	Tool	Stars	Downloads/wk
#1	SkillSpectorSkillSpector is NVIDIA's security scanner that checks AI agent skills for vulnerabilities, malicious patterns, and supply-chain risks…	23.7k	–
#2	JudgevalJudgeval is an open-source Python SDK for tracing and evaluating LLM-powered agents with prompt-based judges and online monitoring.	1.1k	80.2k
#3	A.I.G (AI-Infra-Guard)A full-stack AI Red Teaming platform securing AI ecosystems via Agent Scan, Skills Scan, MCP scan, AI Infra scan and LLM jailbreak…	7.2k	–
#5	OpikOpik is an open-source platform from Comet for tracing, evaluating, and monitoring LLM applications, RAG systems, and AI agents.	23.1k	709.1k
#6	promptfoopromptfoo is a CLI and library for evaluating and red-teaming LLM-based apps, including prompts, agents, and RAG pipelines.	26.7k	1.2M
#7	garakgarak is a command-line LLM vulnerability scanner from NVIDIA that probes models for jailbreaks, prompt injection, data leakage,…	9.8k	12.9k

All LLM evals and observability tools, ranked

Recent Langfuse releases

v4.67.0, Oct 21, 2026
v4.66.0, Oct 20, 2026
v4.65.0, Oct 20, 2026
v4.64.0, Oct 19, 2026
v4.63.0, Oct 16, 2026

Langfuse FAQ

Is Langfuse free and open source?

Langfuse's code is on GitHub (langfuse/langfuse), but the repository doesn't declare a standard open-source license. Check its license file before using it commercially. Its website links to a pricing page: https://langfuse.com/pricing.

How do I install Langfuse?

Run pip install langfuse openai. It needs Node.js >=18. The install section above has the other options from the project's README.

Is Langfuse still maintained?

Yes. The langfuse/langfuse repository was last updated on Oct 21, 2026, and the latest release, v4.67.0 on GitHub, came out on Oct 21, 2026.

How popular is Langfuse?

Langfuse has 36.8k GitHub stars as of Nov 1, 2026. It gained 414 new stars in the week of Oct 25–31, 2026, against a 12-week average of 353 a week. Its main package had about 9.8M downloads that week, +9% against its own 4-week average. That made it #4 among LLM evals and observability tools on the Devtoolwatch Board.

Langfuse vs LangChain: how do they compare?

Langfuse and LangChain are both tracked by Devtoolwatch. Langfuse has 36.8k GitHub stars and LangChain has 148.8k. In the week of Oct 25–31, 2026, Langfuse's main package had 9.8M downloads and LangChain's had 41.1M. LangChain review.

Langfuse vs MLflow: how do they compare?

Langfuse and MLflow are both tracked by Devtoolwatch as LLM evals and observability tools. Langfuse has 36.8k GitHub stars and MLflow has 28.6k. In the week of Oct 25–31, 2026, Langfuse's main package had 9.8M downloads and MLflow's had 5.2M. MLflow is an open source AI engineering platform for tracing, evaluating, monitoring, and managing agents, LLM applications, and ML models. MLflow review.

What are the best alternatives to Langfuse?

Other LLM evals and observability tools on the Devtoolwatch Board in the week of Oct 25–31, 2026, by momentum: SkillSpector (#1), Judgeval (#2), A.I.G (AI-Infra-Guard) (#3), Opik (#5), promptfoo (#6). The alternatives table above has their numbers.

Is Langfuse open source?

Langfuse is open source. The repository is MIT licensed except for the ee folders, and the Python SDK and npm package are MIT licensed.

How do I use Langfuse locally?

Langfuse runs locally with Docker Compose. Clone the repository with git clone --depth=1, change into the langfuse directory, and run docker compose up. The README says this takes about 5 minutes.

How do I use Langfuse with LangChain?

Langfuse instruments LangChain applications automatically when you pass a Langfuse callback handler to the LangChain application. This works in both Python and JS/TS.

Does Langfuse support LLM-as-a-judge evaluation?

Langfuse supports LLM-as-a-judge evaluators, plus code evaluators, user feedback, manual labeling, and custom pipelines. You can run evaluators on production data or during experiments.

Is Langfuse related to LangChain?

The sources do not describe Langfuse as part of LangChain. Langfuse integrates with LangChain as a framework. Since January 2026, Langfuse has been part of ClickHouse.

How do I use the Langfuse API?

Langfuse exposes a public API with an OpenAPI spec and a Postman collection. It also provides typed SDKs for Python and JS/TS. Teams use the API to build custom LLMOps workflows.

Sources

Langfuse README badge

Maintainers can show the tool's weekly Board rank in their README:

[![Devtoolwatch](https://index.devtoolwatch.com/badge/langfuse.svg)](https://index.devtoolwatch.com/tools/langfuse/)

How this page is made: Devtoolwatch writes each review from the project's README, releases, package metadata and website, with an AI model, then checks every claim against the source it cites and drops what the sources don't support. The numbers come from Devtoolwatch's daily GitHub snapshots, GitHub's stargazer history, and the npm and PyPI download APIs. Spotted an error? Email [email protected].