Laminar AI

Laminar is an open-source observability platform purpose-built for AI agents. It combines tracing, evaluation, and agent-behavior monitoring in a single system, with a Rust backend built for high-volume ingestion.

Tracing - An OpenTelemetry-native SDK that instruments your agent in a single line of code. Automatic tracing for Vercel AI SDK, Browser Use, Stagehand, LangChain, OpenAI, Anthropic, Gemini, and more.

Signals - Describe any agent behavior you want to track in plain English, such as "agent is stuck in a loop." Laminar reads every agent run and notifies you in Slack when the behavior occurs.

Evals - An unopinionated, extensible SDK and CLI for running evaluations locally or in CI/CD, plus a UI for visualizing runs and comparing results.

MCP and CLI access - Query traces, spans, metrics, and events with SQL, and let your coding agent investigate and debug issues directly against your trace data.

Dashboards - A dashboard builder for traces, metrics, and events, with support for custom SQL queries.

Data annotation and datasets - Custom data rendering UI for fast annotation and dataset creation for evals.

Performance - Written in Rust, with 20x trace compression for efficient ingestion and storage, a custom realtime engine for viewing traces as they happen, ultra-fast full-text search over span data, and a gRPC exporter for tracing data.