Langfuse
Open-source LLM engineering platform — tracing, evals, prompt management, and metrics.
Acquired Langfuse
Acquired · December 2025
Acquired by ClickHouse in December 2025; Langfuse continues to operate independently as the LLM observability platform.
Acquired by ClickHouse.
What to use instead
Grok
Elon Musk's xAI aims to understand the universe's true nature.
Paid - Inquire
4.92
482
fal.ai
Fastest generative AI platform for developers — 1,000+ image, video, audio, and 3D models with optimized real-time inference. Default home for FLUX, SAM, MuseTalk.
Freemium
4.93
490
Vercel AI SDK
Universal TypeScript SDK from Vercel for building AI apps and agents with multi-model support.
Free
4.93
480
Overview
Langfuse: Open-Source LLM Engineering Platform
Langfuse is the open-source LLM engineering platform — traces, evals, prompt management, and metrics in one integrated stack. Used by 19 of the Fortune 50, ingesting 10+ billion observations per month, with 100,000+ engineers building on it. Now part of ClickHouse, with a continued focus on building the best LLM engineering platform in the open.
Debug AI applications and agents in minutes, spot issues before users do, and collaborate with your team to continuously improve cost, latency, and quality. Langfuse works with any model, any framework, and is based on OpenTelemetry — so it slots into your existing observability stack without lock-in.
Key Features:
- Open-source — Apache 2.0 with full self-hosting
- LLM observability with deep trace visibility
- Prompt management as a first-class concept
- Evaluation framework, including LLM-as-a-Judge
- Metrics for cost, latency, and quality
- Experiments as a first-class concept
- OpenTelemetry-based, model and framework agnostic
- Self-hosting via Docker Compose, Kubernetes, AWS, GCP, Azure
- Used by Khan Academy, Adobe, Twilio, Telus, Intuit, Cisco, Expedia
- 26.4k GitHub stars, 300+ contributors
Ideal Use Case:
Langfuse is ideal for engineering teams building LLM applications who want a single open-source platform for tracing, evals, prompt management, and production metrics. Especially strong for teams that need self-hosting for compliance or data residency.
Why Use Langfuse:
- Fully open source — Apache 2.0
- Self-host on your own infrastructure
- One platform for traces, evals, prompts, and metrics
- OpenTelemetry-native — no framework lock-in
- Used by 19 of the Fortune 50
FAQ
Is Langfuse really free? Yes. Apache 2.0, fully open source. Cloud tiers exist for managed hosting.
Can I self-host? Yes — Docker Compose, Kubernetes (Helm), AWS, GCP, and Azure Terraform modules provided.
Which frameworks does Langfuse support? Any. OpenTelemetry-based with native integrations for OpenAI, LangChain, and the major SDKs.
Who is using Langfuse? 19 of the Fortune 50, including Khan Academy, Adobe, Twilio, Telus, Intuit, Cisco, and Expedia.
tl;dr:
Langfuse is the open-source LLM engineering platform — traces, evals, prompt management, and metrics, all in one. Self-host or use Langfuse Cloud. 26.4k GitHub stars, used by 19 of the Fortune 50.
Related
Looking for more options? Browse the AI Infrastructure directory or read our best AI infrastructure tools listicle. Langfuse is also tracked on Crunchbase.
Why Use Langfuse
FAQ
