Repolode
Submit
lmnr-ai

lmnr-ai/lmnr

Open-source observability platform for AI agents, with tracing, evals, signals and dashboards, self-hostable via Docker Compose.

Open-source alternative to:LangSmithLangfuse CloudDatadog

View on GitHubWebsite
Stars
3,428
Forks
250
Open issues
141
Last commit
9 hours ago

What is it?

Laminar is an observability platform built for AI agents. It records agent runs as OpenTelemetry-based traces, so developers can see each LLM call, tool call and span in one place. The Python and TypeScript SDKs instrument common libraries such as OpenAI, Anthropic, LangChain, Vercel AI SDK and Browser Use with a single initialization call, and a decorator or wrapper captures function inputs and outputs.

Beyond tracing, Laminar covers evaluation and monitoring. Signals let users describe a behavior in plain English, such as an agent stuck in a loop, and send Slack alerts when matching runs occur. An SDK and CLI run evals locally or in CI/CD, and a UI compares results. Datasets and an annotation interface support building eval data. MCP and CLI access let coding agents query traces with SQL.

The backend is written in Rust and the project can be self-hosted with docker compose, with a managed cloud version also offered. The repository includes setup options for several LLM providers and for custom Postgres schemas.

Key features

  • 01Automatic tracing for OpenAI, Anthropic, LangChain, Vercel AI SDK and other libraries with one line of setup.
  • 02Plain-English signals that alert to Slack when agent behavior matches a description.
  • 03SDK and CLI for running evals locally or in CI/CD, with a UI to compare results.
  • 04SQL querying of traces, spans, metrics and events, including MCP access for coding agents.
  • 05Dashboard builder with custom SQL queries and dataset annotation tools.
  • 06Self-hosting via docker compose.

Use cases

  • Debugging production AI agents by inspecting each LLM and tool call in a trace.
  • Running regression evals for agents in a CI/CD pipeline.
  • Getting alerted when agents loop or behave unexpectedly.
  • Self-hosting an agent observability stack to keep trace data in your own infrastructure.

Quick start

  1. 1

    Clone the repository

    git clone https://github.com/lmnr-ai/lmnr
  2. 2

    Enter the repository directory

    cd lmnr
  3. 3

    Start the services with Docker Compose

    docker compose up -d
  4. 4

    Install the TypeScript SDK in your project

    npm add @lmnr-ai/lmnr

Things to know

  • Self-hosted deployments collect anonymized usage telemetry unless you set LAMINAR_TELEMETRY_DISABLED=true.
  • Frontend AI features and AI workers need an LLM provider API key or AWS credentials to be configured.
  • The Docker Compose quickstart is a lightweight stack; the README recommends the full compose file or managed platform for production.

Topics

#agent-observability#llm-observability#tracing#evals#opentelemetry#self-hosted#typescript#rust#agents#ai

FAQ

What is lmnr?
Open-source observability platform for AI agents, with tracing, evals, signals and dashboards, self-hostable via Docker Compose. Laminar is an observability platform built for AI agents. It records agent runs as OpenTelemetry-based traces, so developers can see each LLM call, tool call and span in one place. The Python and TypeScript SDKs instrument common libraries such as OpenAI, Anthropic, LangChain, Vercel AI SDK and Browser Use with a single initialization call, and a decorator or wrapper captures function inputs and outputs.
Is lmnr free to use? What license does it have?
lmnr is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is lmnr still maintained?
Actively maintained. The last commit to lmnr was on Oct 21, 2026.
What can I use lmnr for?
Debugging production AI agents by inspecting each LLM and tool call in a trace. Running regression evals for agents in a CI/CD pipeline. Getting alerted when agents loop or behave unexpectedly. Self-hosting an agent observability stack to keep trace data in your own infrastructure.
How do I install lmnr?
Clone the repository: git clone https://github.com/lmnr-ai/lmnr Enter the repository directory: cd lmnr Start the services with Docker Compose: docker compose up -d Install the TypeScript SDK in your project: npm add @lmnr-ai/lmnr
Where can I get lmnr?
The source code and installation instructions are on GitHub: https://github.com/lmnr-ai/lmnr

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects