Skip to the content
Know your tech before you buy it. Makers: get listed from $49/mo

Laminar

laminar.sh

  • Free plan
  • Web
  • API
  • Self-hosted
7.4 Club score Visit laminar.sh

What Laminar is

Laminar is an open-source observability and debugging platform for AI agents built around OpenTelemetry. It traces LLM calls, tool use, custom functions, and sub-agents, presenting runs as readable transcripts. Signals turn plain-language descriptions of outcomes or failures into structured events across traces. Its debugger can replay a run from cache up to the point being examined, while evaluations can run against datasets locally or in CI to catch regressions. Other tools include realtime traces, full-text search, raw SQL access through the UI, CLI, or MCP, and custom dashboards. Documented integrations include OpenAI, Anthropic, Gemini, Cohere, LangChain, Pydantic AI, Claude Code, Codex, and Playwright. Teams can use managed cloud or self-host, including deployments on AWS or GCP with Helm charts. The Free plan costs nothing and includes 1 GB of data, seven-day retention, one project, and one seat. Starter is $30 a month; Pro is $150 a month, and Enterprise pricing is custom. Laminar states HIPAA and SOC 2 Type II compliance; its PII redaction is best-effort and should not be the sole safeguard.

Who it suits

Laminar suits teams building or operating AI agents that need traces, debugging, and dataset evaluations. It offers both managed cloud and self-hosted deployment options.

Good to know, and what to watch

  • Traces LLM calls, tools, functions, and sub-agents
  • Replays runs from cache for debugging
  • Runs evaluations locally or in CI
  • Supports managed cloud and self-hosting
  • Free plan has seven-day retention
  • Free plan is limited to one project and one seat
  • PII redaction is best-effort
  • Starter costs $30 a month

Enterprise

—

Free

Free

Starter

$30/mo

Pro

$150/mo

Free plan

The verdict

Laminar combines agent tracing with replay debugging, evaluations, and data-query tools. Its free plan is limited in retention and scale, while the listed paid plans add capacity and support options.

What the maker publishes

Company
Y Combinator lists Laminar as founded in 2024 and located in London, United Kingdom. source
Data tools
The platform offers realtime traces, full-text search, raw SQL access through the UI, CLI or MCP, and custom dashboards. source
Debugger
The debugger can replay a run from cache up to the point being tested, with prior calls served from cache. source
Deployment
Laminar offers managed cloud and self-hosting; its site says self-hosted deployments can run on AWS or GCP with Helm charts. source
Evaluations
Users can run evaluations against datasets locally or in CI to catch regressions. source
Integrations
Documented integrations include OpenAI, Anthropic, Gemini, Cohere, LiteLLM, Vercel AI SDK, LangChain, Pydantic AI, Claude Code, Codex, Browser Use, Stagehand, Playwright, and others. source
Privacy
Privacy Mode defaults to off on Free and Starter, on for Pro and higher, and on and unchangeable for accounts covered by a signed DPA; when enabled, Customer Data is not used to train or improve its machine-learning models. source
Privacy limits
The privacy notice says PII redaction is best-effort and applies to model input/output content, so customers should not rely on it as their only safeguard. source
Product
Laminar is an open-source, OpenTelemetry-native observability and debugging platform built for AI agents. source
Security
Laminar states it is HIPAA and SOC 2 Type II compliant and offers server-side PII redaction on eligible plans. source
Signals
Signals let users describe outcomes and failures in plain language and extract structured events across traces. source
Support
The listed support options are community support on Free, email on Starter, Slack on Pro, and dedicated support for Enterprise. source
Tracing
It traces LLM calls, tool use, custom functions, and sub-agents, and presents traces as readable transcripts. source

Checked against the maker's pages on 3 Oct 2026.