Jan is a free, source-available and fair code licensed AI Inference Platform. We help enterprises, small businesses and hobbyists to self-host AI on their own infrastructure efficiently, to protect their data, lower costs, and put powerful AI capabilities in the hands of users.
- Web, Mobile and APIs
- LLMs and Generative Art models
- AI Catalog
- Model Installer
- User Management
- Support for Nvidia, Apple Silicon, CPU architectures
- Nvidia GPUs
- Apple Silicon
- CPU architectures (not recommended)
Jan offers an Docker Compose deployment that automates the setup process.
# Install and update Nvidia Docker Container Runtime
nvidia-smi
# Docker Compose up
docker compose up| Service (Docker) | URL |
|---|---|
| Jan Web | localhost:1337 |
| Jan API | localhost:1337/api |
| Jan API (Swagger) | localhost:1337/api/swagger |
| Jan Docs | localhost:1337/docs |
| Keycloak Admin | localhost:1337/users |
| Grafana Dashboard | localhost:1337/grafana |
- Architecture Diagram
- Keycloak Community (Apache-2.0)
- KrakenD Community Edition (Apache-2.0)
Jan is a monorepo that pulls in the following submodules
├── docker-compose.yml
├── mobile-client
├── web-client
├── app-backend
├── inference-backend
├── docs # Developer Docs
├── adrs # Architecture Decision Records