Metrics
Langfuse metrics derive actionable insights from observability and evaluation traces.
Metrics can be sliced and diced via the customizable dashboards and the metrics API. Set up alerts to get notified when a metric crosses a threshold.
For how to decide which metrics are worth tracking, and how to turn what you see into concrete fixes, read Monitoring in the Langfuse Academy and its chapter on error analysis.
Features
Metrics & Dimensions
Metrics:
- Quality is measured through user feedback, model-based scoring, human-in-the-loop scored samples or custom scores via SDKs/API (see scores). Quality is assessed over time as well as across prompt versions, LLMs and users.
- Cost and Latency are accurately measured and broken down by user, session, geography, feature, model and prompt version.
- Volume based on the ingested traces and tokens used.
Dimensions:
- Trace name: differentiate between different use cases, features, etc. by adding a
namefield to your traces. - User: track usage and cost by user. Just add a
userIdto your traces (docs). - Tags: filter different use cases, features, etc. by adding tags to your traces.
- Release and version numbers: track how changes to the LLM application affected your metrics.
For an exact definition, please refer to the metrics API docs.
Was this page helpful?
Last edited
