Self-Hosting Overview - LangWatch
LangWatch Overview
LangWatch is an open-source LLM Ops platform for evaluation, observability, and optimization of AI agents and pipelines. You can self-host LangWatch for full data sovereignty, regulatory compliance (GDPR, SOC 2), or air-gapped environments. The self-hosted edition runs the same software that powers LangWatch Cloud — there is no separate “community” or “enterprise” build.
Get the entire LangWatch platform running with a single Helm install:
helm repo add langwatch https://langwatch.github.io/langwatch
helm repo update
helm pull langwatch/langwatch --untar
helm install langwatch ./langwatch \
--namespace langwatch --create-namespace \
-f langwatch/examples/values-local.yaml
See Kubernetes (Helm) for full setup instructions.
Architecture at a Glance
| Component | Description |
|---|---|
| OTel / REST | Data collection and transmission |
| Your LLM App | The application hosting your model |
| LangWatch App | The LangWatch user interface |
| AI Gateway | Optional component for managing traffic |
| LLM Providers | e.g., OpenAI, Anthropic, Bedrock |
| Redis | Caching layer |
| LangWatch Workers | Asynchronous processing |
| ClickHouse | Analytical database |
| LangEvals | Evaluation framework |
| LangWatch NLP | Natural Language Processing capabilities |
| PostgreSQL | Relational database for state |
| S3 | Object storage |
See Architecture & Infrastructure for a detailed breakdown of networking, scaling, and storage tiers.
Deployment Models
LangWatch Cloud
Fully managed by the LangWatch team. No infrastructure to manage, sign up and start sending traces in minutes. Sign up at langwatch.ai
Self-Managed
Deploy the complete LangWatch stack on your own infrastructure, AWS, Azure, GCP, or bare metal. You manage everything: compute, databases, and storage. Full data sovereignty. Get started with Kubernetes
Cloud Enterprise
LangWatch manages the application. Your data lives on exclusive, dedicated instances deployed in your preferred cloud region.
Hybrid
LangWatch manages the control plane while you provide the data plane, ensuring your trace data never leaves your network. Learn more
At a Glance
| Cloud | Self-Managed | Cloud Enterprise | Hybrid | |
|---|---|---|---|---|
| You manage infrastructure | Yes | |||
| You manage data storage | Yes | Yes | Yes | |
| LangWatch manages app | Yes | Yes | Yes | |
| Data stays in your network | Yes | Yes | Yes | |
| Setup time | Minutes | Hours | Days | Days |
Quick Start
Choose the deployment method that fits your environment:
- Docker Compose: Quick local setup with Docker.
- Kubernetes (Helm): Production deployment on any Kubernetes cluster.
- Local Kubernetes: Test the Helm chart locally with Kind.
What’s New in v3
LangWatch v3 is a major architecture upgrade. If you are migrating from v2, review the deployment guides for updated requirements.
- ClickHouse replaces Elasticsearch as the primary data store.
- Event-sourcing architecture ensures reliable processing of traces.
- S3 cold storage tiering automatically moves older data to reduce costs.
- Native ClickHouse backup/restore simplifies disaster recovery.
- Auto-tuned ClickHouse optimizes resource allocation.
- Composable Helm chart overlays allow customizable deployments.
- Deep OpenTelemetry integration for infrastructure monitoring.
- AI Gateway sub-chart allows for LLM traffic management with virtual keys. See AI Gateway → Self-hosting.
Enterprise Features
For Cloud Enterprise or Hybrid deployments, contact the LangWatch team to discuss your requirements. Enterprise capabilities include:
- SSO, SCIM: Integrate with your identity provider for user provisioning.
- Role-based access control: Fine-grained permissions across projects.
- Audit logs: Full visibility into system activities.
- Priority support: Dedicated engineering assistance.