This endpoint is in preview and may be modified or removed at any time.
To use this endpoint, add preview=true to the request query parameters.
Updates the runtime configuration of the live deployment. The deployment will apply the new configuration to the running replicas.
Third-party applications using this endpoint via OAuth2 must request the following operation scope: api:models-write.
objectobjectThe compute resource configuration for the deployment.
objectThe replaced LiveDeployment
stringThe Resource Identifier (RID) of a Live Deployment.
objectThe currently deployed model version.
stringoptionalThe model branch this deployment tracks. Present for direct deployments that follow the latest model version on a branch; absent for deployment types that are not branch-scoped.
objectThe compute resource configuration for the deployment.
objectThe current operational status of the deployment.
1
2
3
4
5
curl -X PUT \
\t-H "Content-type: application/json" \
\t-H "Authorization: Bearer $TOKEN" \
"https://$HOSTNAME/api/v2/models/liveDeployments/ri.foundry-ml-live.main.live-deployment.f351c142-0e4c-4b12-adc2-6e1539737ae9?preview=true" \
-d '{"runtimeConfiguration":{"minReplicas":1,"maxReplicas":3,"cpu":1.0,"memory":"256MiB","threadCount":32}}'1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
{
"runtimeConfiguration": {
"minReplicas": 1,
"maxReplicas": 3,
"cpu": 1,
"memory": "256MiB",
"threadCount": 32
},
"modelVersion": {
"modelRid": "ri.models.main.model.f351c142-0e4c-4b12-adc2-6e1539737ae9",
"modelVersionRid": "ri.models.main.model-version.adf94926-c3ac-41ea-beb2-4946699d08ee"
},
"rid": "ri.foundry-ml-live.main.live-deployment.f351c142-0e4c-4b12-adc2-6e1539737ae9",
"branch": "master",
"status": {
"state": "ACTIVE",
"isReady": true
}
}| Error Name | ||
|---|---|---|
Thread | Error Code | INVALID_ARGUMENT |
| Status Code | 400 | |
| Description | The specified thread count exceeds the maximum allowed value. | |
| Parameters | maxThreadCount, providedThreadCount | |
Invalid | Error Code | INVALID_ARGUMENT |
| Status Code | 400 | |
| Description | The GPU count is invalid. The GPU count must be between 1 and the maximum allowed for the requested GPU type. | |
| Parameters | providedGpuCount, maxGpuCount | |
Gpu | Error Code | INVALID_ARGUMENT |
| Status Code | 400 | |
| Description | The requested GPU type is not available. Use a GPU type that is available in the deployment's resource queue. | |
| Parameters | requestedGpuType, availableGpuTypes | |
Live | Error Code | NOT_FOUND |
| Status Code | 404 | |
| Description | The specified live deployment was not found. | |
| Parameters | liveDeploymentRid | |
Unsupported | Error Code | INVALID_ARGUMENT |
| Status Code | 400 | |
| Description | The Live Deployment type is not supported by the API. | |
| Parameters | | |
Replace | Error Code | PERMISSION_DENIED |
| Status Code | 403 | |
| Description | Could not replace the LiveDeployment. | |
| Parameters | liveDeploymentRid | |
See Errors for a general overview of errors in the platform.