Cette page est affichée en anglais. Une traduction française est en cours.
ConfigsMar 31, 2026·2 min de lecture

Langfuse — Open Source LLM Observability

Langfuse is an open-source LLM engineering platform for tracing, prompt management, evaluation, and debugging AI apps. 24.1K+ GitHub stars. Self-hosted or cloud. MIT.

Introduction

Langfuse is an open-source LLM engineering platform that helps teams collaboratively develop, monitor, evaluate, and debug AI applications. With 24,100+ GitHub stars and MIT license, Langfuse provides LLM observability with detailed tracing of calls, retrieval, embeddings, and agent actions. It includes centralized prompt management with version control, evaluation pipelines (LLM-as-judge, user feedback, manual labeling), dataset testing, an interactive playground, and SDKs for Python and JavaScript. Self-host in minutes with Docker, Kubernetes, or deploy on AWS/Azure/GCP.

Best for: Teams building production AI apps who need observability, prompt management, and evaluation Works with: Claude Code, OpenAI Codex, Cursor, Gemini CLI, Windsurf Integrations: OpenAI, LangChain, LlamaIndex, Haystack, Vercel AI SDK


Key Features

  • LLM tracing: Track every call, retrieval, embedding, and agent action
  • Prompt management: Version control, collaboration, server/client caching
  • Evaluations: LLM-as-judge, user feedback, manual labeling, custom pipelines
  • Datasets: Benchmark testing for continuous improvement
  • Playground: Interactive prompt testing and iteration
  • Self-hosted: Docker, Kubernetes, Terraform (AWS/Azure/GCP)

FAQ

Q: What is Langfuse? A: Langfuse is an LLM observability platform with 24.1K+ stars for tracing, prompt management, and evaluation. Self-host with Docker or use cloud. MIT licensed.

Q: How do I install Langfuse? A: pip install langfuse openai. Swap from openai import OpenAI to from langfuse.openai import openai for automatic tracing.


🙏

Source et remerciements

Created by Langfuse. Licensed under MIT. langfuse/langfuse — 24,100+ GitHub stars

Discussion

Connectez-vous pour rejoindre la discussion.
Aucun commentaire pour l'instant. Soyez le premier à partager votre avis.

Actifs similaires