API Reference
Programmatically manage your AI infrastructure. Provision NVIDIA B200, H100, and A100 instances, manage NVMe volumes, and configure VPC networks using our secure REST API.
Authentication
All API requests must be authenticated via a Bearer token in the HTTP Authorization header. You can generate API keys from your RSP Cloud console.
⚠️ Keep your API keys secure. Never commit them to public repositories.
Authorization: Bearer rsp_live_xYz123aBc456...
Compute Instances
Provisions a new GPU compute instance. Billing begins as soon as the instance enters the running state.
Parameters (JSON)
curl -X POST https://api.rsplatforms.com/v1/compute/instances \
-H "Authorization: Bearer $RSP_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"gpu_type": "H100",
"node_count": 8,
"image_id": "ubuntu-22.04-pytorch-latest"
}'
{
"id": "inst_9a8b7c6d5e4f",
"status": "provisioning",
"gpu_type": "H100",
"node_count": 8,
"created_at": "2026-06-26T12:35:00Z"
}
Returns a list of all your active and stopped compute instances.
curl -X GET https://api.rsplatforms.com/v1/compute/instances \
-H "Authorization: Bearer $RSP_API_KEY"
Permanently terminates an instance. All local data on the instance NVMe drives will be destroyed. Billing stops immediately.
curl -X DELETE https://api.rsplatforms.com/v1/compute/instances/inst_9a8b7c \
-H "Authorization: Bearer $RSP_API_KEY"
Rate Limits & Enterprise Quotas
Standard API requests are limited to 1000 requests / minute per IP address. GPU provisioning quotas (number of simultaneous B200/H100 instances) are determined by your account tier.