Get deep insights and evaluate your LLM application data
Langfuse is an open-source LLM engineering platform that helps teams collaboratively debug, analyze, and iterate on their LLM applications. All platform features are natively integrated to accelerate the development workflow. Langfuse is open. It works with any model, any framework, allows for complex nesting, and has open APIs to build downstream use cases. Docs: https://langfuse.com/docs Github: https://github.com/langfuse/langfuse
Dashboards support filtering and grouping by trace name, user ID, tags, and custom metadata, enabling teams to create focused views for different use cases, features, or user segments. This helps product managers and engineers understand performance variations across different parts of the application.
Custom dashboards allow users to plot evaluation results, compare scores across prompt versions and LLM models side-by-side, and track quality trends over time. This helps teams identify regressions before shipping changes and measure the impact of prompt or model iterations.
Langfuse Custom Dashboards enable teams to create customizable dashboards that track cost, latency, and quality metrics across traces, users, sessions, and prompt versions. Users can slice data by custom dimensions and set up automated alerts on operational thresholds, providing visibility into production LLM system performance.
Langfuse dashboards support exports to PostHog and Mixpanel, and provide S3 blob storage exports and REST APIs for querying metrics and observations. This allows teams to integrate LLM metrics into existing analytics workflows and build custom reports in external BI platforms.