Open Laboratory Docs
Run open models and AI apps on your own GPU — launch Laboratory OS anywhere with one command.
Get started
Jump straight into Quick Start.
Getting Started
- Quick StartGet your Open Laboratory container running in minutes.
- Core ConceptsUnderstand how Open Laboratory's container, persistence, tunnels, subdomains, and security model fit together.
- Install DependenciesInstall Docker, NVIDIA drivers, and the NVIDIA Container Toolkit on a Linux machine.
- Run Laboratory OSDocker run reference — image tags, options, CPU-only mode, and a sample Docker Compose file.
Guides
- Video Generation with Wan 2.2 and ComfyUIBuild a text-to-video pipeline using Wan 2.2 inside a ComfyUI workflow — from model install to first render.
- Private LLM API with Qwen3 and vLLMDeploy a production-grade OpenAI-compatible endpoint serving Qwen3 at full GPU throughput — no token costs, no rate limits.
- Autonomous Coding Agent with OpenCodeRun OpenCode inside a sandboxed Laboratory desktop — full filesystem access, terminal, and live supervision from your browser.
Tutorials
Reference
Deployment
- RunpodDeploy Laboratory OS on a Runpod GPU pod.
- Vast.aiDeploy Laboratory OS on a Vast.ai GPU instance.
- Amazon Web ServicesDeploy Laboratory OS on an AWS EC2 GPU instance.
- Google Cloud PlatformDeploy Laboratory OS on a GCP Compute Engine GPU VM.
- Microsoft AzureDeploy Laboratory OS on an Azure GPU VM.
- PaperspaceDeploy Laboratory OS on a Paperspace GPU machine.
- DigitalOceanDeploy Laboratory OS on a DigitalOcean GPU Droplet.
- TensorDockDeploy Laboratory OS on a TensorDock GPU server.
- VultrDeploy Laboratory OS on a Vultr Cloud GPU instance.
- NebiusDeploy Laboratory OS on a Nebius GPU VM.
- VerdaDeploy Laboratory OS on a Verda GPU instance.
- LambdaDeploy Laboratory OS on a Lambda GPU instance.
- OblivusDeploy Laboratory OS on an Oblivus GPU instance.
- JarvisLabsDeploy Laboratory OS on a JarvisLabs GPU instance.
- Oracle CloudDeploy Laboratory OS on an Oracle Cloud GPU instance.
- Akamai (Linode)Deploy Laboratory OS on an Akamai Cloud (Linode) GPU instance.
- Thunder ComputeDeploy Laboratory OS on a Thunder Compute GPU instance.
- Alibaba CloudDeploy Laboratory OS on an Alibaba Cloud GPU instance.
- Nova CloudDeploy Laboratory OS on a Nova Cloud GPU instance.
- SaladDeploy Laboratory OS on Salad Cloud's distributed GPU network.