Inference Instances
Start inference deployment
This operation initializes an inference deployment after it was stopped, making it available to handle inference requests again. The instance will launch with the minimum number of replicas defined in the scaling settings.
- If the minimum replicas are set to 0, the instance will initially start with 0 replicas.
- It will automatically scale up when it receives requests or SQS messages, according to the configured scaling rules.
POST
Authorizations
API key for authentication.
Path Parameters
Project ID
Examples:
1
Inference instance name.
Required string length:
4 - 30
Examples:
"my-instance"
Response
204
No Content