Langfuse
Langfuse enhances LLM applications with detailed observability, offering logging, tracing, and insights for improved performance monitoring.
v4.76.0 · Stable•MIT•Since May 2023•Has commercial version
Overall rating
activity, maturity, stars
9.1
/ 10
GitHub stars
37k
Contributors
229
Last commit
13h ago
Project age
3 years
About
Langfuse is a platform designed for managing, analyzing, and improving AI-driven applications. It provides tools for observability, debugging, and fine-tuning large language models (LLMs). By tracking interactions and model performance, Langfuse helps developers identify and resolve issues, optimize workflows, and enhance application quality. It supports seamless integration with various AI systems, ensuring efficient monitoring and diagnostics.
Key Features
- LLM Observability: Instrument your app and start ingesting traces to Langfuse
- Langfuse UI: Inspect and debug complex logs
- Prompt Management: Manage, version and deploy prompts from within Langfuse
- Prompt Engineering: Test and iterate on your prompts with the LLM Playground
- LLM Analytics: Track metrics (cost, latency, quality) and gain insights from dashboards & data exports
- LLM Evaluations: Collect and calculate scores for your LLM completions
- Run and LLM-as-a-Judge within Langfuse
- Collect user feedback
- Manually score LLM outputs in Langfuse
- Experiments: Track and test app behaviour before deploying a new version
- Datasets let you test expected in and output pairs and benchmark performance before deploying
- Track versions and releases in your application