Langfuse
Open-source observability and evaluation platform for LLM applications
What is Langfuse?
Langfuse is an open-source AI engineering platform for teams building, monitoring, and improving LLM applications. It brings tracing, prompt management, evaluations, and analytics dashboards into one product.
The platform is designed to support the path from a prototype through production use. Teams can use production signals to inspect behavior, test changes, and evaluate agent quality over time.
Its observability tools capture LLM calls, tool invocations, and retrieval steps in hierarchical traces. Developers can explore those records by dimensions such as user, session, cost, latency, and custom metadata.
Langfuse also provides prompt management and a playground for testing prompts on production inputs and comparing models. It is relevant for engineering teams that want a connected workflow for operating and iterating on AI features.
Langfuse Features
LLM tracing and observability
Langfuse records hierarchical traces for LLM calls, tool invocations, and retrieval steps. Those traces can be filtered by operational attributes including user, session, cost, latency, and custom metadata.
Prompt management and testing
Langfuse separates prompts from application code and supports deployments and rollbacks. Its playground is intended for testing prompts against real production inputs and comparing models side by side.
Evaluation and experimentation workflow
Langfuse connects tracing, monitoring, datasets, experiments, and evaluation in a continuous workflow. Teams can use production signals to understand behavior, test improvements, and assess agent quality as they make changes.
Integrations
Pricing
Pricing not verified
User reviews
No reviews yet. Be the first to review this tool.
Log in to write a review.