Open Source AI
Engineering Platform
Trace and evaluate AI Agents. Collaborate with your team to continuously improve quality, cost and latency of your application.
Gain deep visibility into your traces
Track model cost and latency
Improve your prompts
Evaluate model outputs automatically
Collaborate on human reviews
Iterate with structured experiments
Auto-advance is active. Press Escape to pause auto-advance.
Launch, observe, improve — repeat.
Langfuse helps you ship AI Agents/Products from prototype to production and beyond. Once in production we power your continuous improvement loop using production data to make your agents and LLM applications ever more powerful.
The full LLM engineering loop
Langfuse brings observability, prompts, evals, experiments, and human annotation into one connected workflow — so you can move from prototype to production and keep improving with real usage data. Hover any part of the diagram to learn more.
Observability
Trace every LLM call, agent step, and tool use. Debug issues in production with full context.
Evals
Run LLM-as-a-judge, model-based, and custom evaluations on traces automatically.
Experiments
Test prompt changes and model swaps on datasets before shipping to production.
Prompts
Manage, version, and deploy prompts without code changes. A/B test in production.
Human Annotation
Queue traces for human review. Collect ground truth labels to improve your evals.
All the tools, one integrated platform.
One integrated platform to trace, manage prompts, evaluate, and experiment from prototype to production scale.
Hierarchical traces capture every LLM call, tool invocation, and retrieval step.
Filter by user, session, cost, latency, and more.
Track costs and latency across models and deployments.
Break down spend by user, model, and time period.
Iterate on prompts and model configs in the browser.
Test changes against real production traces before deploying.
Trusted by teams at