lmnr-ai/lmnr

Health v2 · maintenance only

Adoption not part of the score

3198 stars · 227 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Laminar is an open-source, OpenTelemetry-native observability platform purpose-built for AI agents, offering tracing, plain-language failure detection (Signals), evaluations, datasets, and dashboards. It is written in Rust for high performance, can be self-hosted via Docker Compose, and provides SDKs, a CLI, and an MCP server for querying trace data.

Use cases

• trace llm calls and tool calls in my ai agent
• detect when my agent gets stuck in a loop and alert me in slack
• run llm evals in ci/cd against datasets
• self-host an llm observability platform
• debug agent failures across thousands of traces
• query agent traces with sql from my coding agent
• build evaluation datasets from production traces

When to choose

• you are building or operating LLM agents in production and need tracing, failure detection, and evals in one platform
• you want OpenTelemetry-native tracing with auto-instrumentation for LangChain, Vercel AI SDK, OpenAI, Anthropic, and more
• you need to self-host observability infrastructure with high-performance ingestion and trace compression

When to avoid

• you only need generic application APM without LLM/agent-specific features
• you want a fully managed zero-maintenance solution and cannot run Docker services
• your project is not LLM/agent-based and would not benefit from trace analysis of agent runs

Facets

service · maturity active

monitoring tracing analytics alerting mcp cli sdk large-language-models monitoring developer-tools self-hosted self-hosted cli rust llm-observability opentelemetry llm-evaluation agent-tracing llmops clickhouse signals evals ai-agents docker web-server nodejs typescript

Member repositories

For agents

markdown · JSON · MCP: product_card(name="lmnr-ai/lmnr")