Base URL
Authentication
All API requests (except/v1/health) require authentication using a Bearer token.
Required Headers
| Header | Required | Description |
|---|---|---|
Authorization | Yes | Bearer token (JWT from Cognito or API token) |
X-Organization-Id | Yes* | Your organization ID (*not required for /v1/health) |
X-Request-Id | No | Optional request ID for tracing |
Response Format
All successful responses follow a consistent structure:Error Responses
Errors return an appropriate HTTP status code with details:Authorization Model
Access to resources is controlled by your role within the organization:| Role | Access Level |
|---|---|
ORG_ADMIN | Full organization access |
TEAM_LEAD | Team-specific access with management capabilities |
MEMBER | Team-specific read access |
Endpoints
Workloads
List, retrieve, and get statistics for GPU workloads
Metrics
Query GPU utilization, memory, temperature, and power metrics
Capacity
Check budget allocations and remaining GPU hours
Health
Service health check endpoint

