Deploy on Runpod
Runpod is a cost-effective GPU cloud that works well for Laboratory OS. Pods start in seconds and you only pay while they’re running.
1. Create a Pod
Log in to runpod.io and click Deploy → GPU Pod.
Select a GPU.
Choosing a GPU: More VRAM lets you run larger models — 16 GB handles most image generation, 24-48 GB covers mid-size LLMs, and 100 GB+ is needed for large LLMs.
2. Configure the Pod
When setting up the pod, choose Custom Image and fill in the fields:
Container Image:
openlaboratoryorg/laboratory-osEnvironment Variables — add one entry:
| Key | Value |
|---|---|
UPLINK_API_KEY | Your account key from openlaboratory.com. The lab connects through it and serves on your *.laboratory.computer URL. |
Volume: Optionally mount a network volume at /data to persist your models and installed apps across pod restarts. Without a volume, data resets each time the pod is recreated.
No inbound ports or firewall rules are needed — Laboratory OS uses an outbound tunnel. You can block all inbound traffic entirely.
3. Deploy and Connect
Click Deploy. The pod starts and the container launches automatically.
Check the container logs for your assigned slug — it appears within the first 30 seconds. Then sign in to your account at:
https://app.laboratory.computerFrom there you'll see your running instance and can open the Laboratory OS desktop. Access is gated by your SSO login rather than a shared token, so only authenticated account holders can connect.
4. Stopping the Pod
Stop the pod from the Runpod dashboard when not in use to pause billing. If you attached a network volume, your models and apps persist. Start it again and Laboratory OS picks up where you left off.
Tip: Runpod’s Spot pods are cheaper but can be interrupted. Use Secure Cloud pods for uninterrupted sessions.