langfuse/langfuse

What it solves

Langfuse is an open-source LLM engineering platform designed to help teams collaboratively develop, monitor, evaluate, and debug AI applications. It addresses the complexity of moving AI apps from prototype to production by providing tools for observability, prompt management, and systematic evaluation.

How it works

Langfuse integrates into AI applications via SDKs (Python, JS/TS) or automated instrumentation for popular frameworks like LangChain, LlamaIndex, and OpenAI. It captures traces of LLM calls, retrieval steps, and agent actions, which are then visualized in a dashboard. The platform also provides a centralized prompt management system with version control and a playground for testing model configurations.

Who it’s for

It is built for AI engineers and development teams who need to monitor their LLM-powered applications in real-time, iterate on prompts collaboratively, and establish a rigorous evaluation pipeline using datasets and benchmarks.

Highlights

• LLM Application Observability: Track and inspect complex logs, user sessions, and internal logic (retrieval, embeddings) through tracing.
• Prompt Management: Centrally manage and version control prompts to iterate without adding application latency.
• Evaluations: Support for LLM-as-a-judge, code evaluators, manual labeling, and custom API-driven evaluation pipelines.
• Datasets: Create test sets and benchmarks for continuous improvement and pre-deployment testing.
• LLM Playground: A dedicated environment to test and iterate on prompts and model configurations directly from traces.
• Flexible Deployment: Available as a managed cloud service or self-hosted via Docker Compose, Kubernetes (Helm), or Terraform templates.