LangWatch is a TypeScript-based platform for evaluating LLMs and testing AI agents, offering end-to-end simulations, evaluation, observability, and prompt management. It provides cloud and self-hosted deployment options, with multiple integrations and OpenTelemetry-based tracing.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
The platform for LLM evaluations and AI agent testing.
How it works
LangWatch supports end-to-end agent simulations against a full stack (tools, state, user simulator, judge) and combines trace, dataset, evaluation, and prompt/model optimization in a single loop. It uses OpenTelemetry for tracing and integrates with multiple frameworks and model providers.
Getting started
Cloud
Create a free account → create a project → get started/ copy your API key.
Local setup
The fastest way to run LangWatch locally — only Node.js required:
npx @langwatch/server
The CLI installs various components into ~/.langwatch/, scaffolds a .env file, then starts all services and opens http://localhost:5560. You can modify evaluators in ~/.langwatch/.env.
LANGWATCH_ENABLE_LANGY
LANGWATCH_ENABLE_PRESIDIO
LANGWATCH_ENABLE_LINGUA
LANGWATCH_ENABLE_LEGACY_EVALUATORS
Changing these values in ~/.langwatch/.env and restarting updates the setup.
Preferred Docker deployment:
git clone https://github.com/langwatch/langwatch.git
cd platform/app
cp platform/app/.env.example platform/app/.env
docker compose up -d --wait --build
Once running, available at http://localhost:5560.
Deployment options
- Docker Compose
- Kubernetes (Helm)
- OnPrem
Local development options include running databases with docker and commands:
docker compose up redis postgres opensearch
make install
make start
Recent releases
Latest Python SDK release: python-sdk@v1.1.0 (2026-08-04). LangWatch v3.9.0 (2026-08-03) with server and gateway updates.
Traction
GitHub stars: 3466.
License
Open-core: Apache-2.0 plus Enterprise extensions; SDKs are MIT.
Security + Compliance
GDPR compliant; ISO 27001 certified; security contact available.
Integrations
Frameworks: LangChain, LangGraph, Vercel AI SDK, Mastra, CrewAI, Google ADK. Model Providers: OpenAI, Anthropic, Azure, Google Cloud, AWS, Groq, Ollama. Platforms: LangFlow, Flowise, n8n.
Getting started – quick links
- Website: https://langwatch.ai
- Docs: https://docs.langwatch.ai
- Discord: https://discord.gg/kT4PhDS2gH
- Self-hosting: https://docs.langwatch.ai/self-hosting/overview