Langfuse

LLM Observability & Tracing

Capture detailed traces of every LLM call, retrieval step, and tool execution with timing, inputs, outputs, and metadata for full request visibility

Prompt Management

Centrally manage, version control, and collaboratively iterate on prompts with server and client-side caching to avoid added latency

Evaluations

Run LLM-as-a-judge evaluations, collect user feedback, perform manual labeling, and build custom evaluation pipelines via APIs and SDKs

LLM Playground

Test and iterate on prompts and model configurations directly, with the ability to jump from traced results into the playground for debugging

Cost & Token Tracking

Monitor usage and costs across all LLM providers with detailed breakdowns by model, trace, and time period

Datasets & Experiments

Create test datasets and run systematic experiments to benchmark prompt and model changes before deploying to production

OpenTelemetry Integration