Waitlist

Queue for unavailable GPU types and get notified when capacity becomes available.

The Waitlist API lets you reserve your place in line for GPU cluster types that are currently at capacity. When a requested GPU type becomes available, you'll be notified via email and webhook — or you can opt into automatic deployment.

Base URL: https://gpu.huddleapis.com/api/v1

How the Waitlist Works

You try to deploy a GPU cluster type, but get a 409 Conflict (no capacity)
Create a waitlist request for that cluster type
System monitors availability as GPU inventory is synced
You get notified when capacity opens up (email + optional webhook)
Deploy manually or let auto-deploy handle it for you

Create a Waitlist Request

curl -X POST -H "X-API-Key: mk_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "cluster_type": "8H200.141S.176V",
    "location": "FIN-01"
  }' \
  https://gpu.huddleapis.com/api/v1/deployments/requests

Response:

{
  "code": "OK",
  "data": {
    "id": "wr_abc123",
    "cluster_type": "8H200.141S.176V",
    "location": "FIN-01",
    "status": "waiting",
    "auto_deploy": false,
    "created_at": "2026-04-03T10:00:00Z",
    "updated_at": "2026-04-03T10:00:00Z"
  }
}

Leave location empty to be notified when the GPU type is available in any region.

Auto-Deploy

If you want to automatically deploy when capacity opens up, pass auto_deploy: true along with a full deployment configuration:

curl -X POST -H "X-API-Key: mk_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "cluster_type": "8H200.141S.176V",
    "location": "FIN-01",
    "auto_deploy": true,
    "config": {
      "hostname": "my-training-cluster",
      "image": "ubuntu-22.04-cuda-12.4-cluster",
      "ssh_key_ids": ["key-id-1"],
      "location": "FIN-01"
    }
  }' \
  https://gpu.huddleapis.com/api/v1/deployments/requests

When the GPU becomes available, the system will automatically create the deployment using your saved config.

Auto-deploy requires sufficient credits

Make sure you have at least $20 in GPU credits at all times while waiting. If your balance is too low when a GPU opens up, auto-deploy will fail.

Waitlist Statuses

Status	Description
`waiting`	Request is in the queue. System is monitoring for availability.
`notified`	GPU capacity opened up. Notification sent via email and webhook.
`deployed`	Auto-deploy was triggered and a deployment was created.
`cancelled`	Request was cancelled by the user.

List Your Requests

curl -H "X-API-Key: mk_your_key" \
  https://gpu.huddleapis.com/api/v1/deployments/requests

Cancel a Request

curl -X DELETE -H "X-API-Key: mk_your_key" \
  https://gpu.huddleapis.com/api/v1/deployments/requests/wr_abc123

Webhook Integration

Subscribe to the gpu.available event in Webhooks to receive real-time notifications when your waitlisted GPU type opens up. This is useful for triggering custom automation pipelines.

Next Steps

Learn about Offers to see what's currently available
Set up Webhooks to get real-time gpu.available notifications
Review Deployments for the full deployment workflow

Waitlist

On this page