Create dedicated endpoint
Provision a new dedicated GPU endpoint. Requires an owner/admin key and enough prepaid balance for the minimum runway. Discover valid model/GPU/region/count combinations from the templates endpoint first.
Authorizations
Your workspace API key, e.g. sk_sovereign_..., sent as Authorization: Bearer <key>.
Body
This body is camelCase, unlike the snake_case loop endpoints. The service validates the model/flavor/GPU/region combo against the deploy templates, prices it, and requires prepaid balance for the first hour.
A template name from GET /v1/dedicated/templates.
A flavor of that template.
At least 1.
At least minReplicas.
Serve a fine-tuned model's weights by artifact id (used when deploying a trained model).
The source fine-tuning job id, when deploying a model trained on this platform.
Response
The created endpoint.
Live-reconciled from the control plane on every list/get.
STARTING, UPDATING, RUNNING, STOPPING, STOPPED, DELETED, WARNING Whether the endpoint is started. Billing runs only while RUNNING and enabled.
Your price per GPU-hour for this configuration (rate frozen at deploy).
Cost accrued since the last metering sweep, at the frozen rate.
Call the endpoint through the gateway as model "dedicated/<routing_key>".
The inference base URL to point your client at.