Purchase a Provisioned Throughput for an Amazon Bedrock model
When you purchase a Provisioned Throughput for a model, you specify the level of commitment for it and the number of model units (MUs) to allot. For MU quotas, see Amazon Bedrock endpoints and quotas in the AWS General Reference. The number of MUs that you can allot to your Provisioned Throughputs depends on the commitment term for the Provisioned Throughput:
-
By default, your account provides you with 2 MUs to distribute between Provisioned Throughputs with no commitment.
-
If you're purchasing a Provisioned Throughput with commitment, you must first visit the AWS support center to request MUs for your account to distribute between Provisioned Throughputs with commitment. After your request is granted, you can purchase a Provisioned Throughput with commitment.
After you purchase the Provisioned Throughput, you can only change the associated model if you select a custom model. You can change the associated model to one of the following:
To learn how to purchase Provisioned Throughput for a model, choose the tab for your preferred method, and then follow the steps:
- Console
-
-
Sign in to the AWS Management Console using an IAM role with Amazon Bedrock permissions, and open the Amazon Bedrock console at
https://console.aws.amazon.com/bedrock/.
-
Select Provisioned Throughput from the left navigation pane.
-
In the Provisioned Throughput section, choose
Purchase Provisioned Throughput.
-
For the Provisioned Throughput details section, do the following:
-
In the Provisioned Throughput name field, enter a name for the Provisioned Throughput.
-
Under Select model, select a base model provider or a custom model category. Then select the model for which to provision throughput.
-
(Optional) To associate tags with your Provisioned Throughput, expand the Tags section and choose Add new tag. For more information, see Tagging Amazon Bedrock resources.
-
For the Commitment term & model units section, do the following:
-
In the Select commitment term section, select the amount of time for which you want to commit to using the Provisioned Throughput.
-
In the Model units field, enter the desired number of model units (MUs). If you are provisioning a model with commitment, you must first visit the AWS support center to request an increase in the number of MUs that you can purchase.
-
Under Estimated purchase summary, review the estimated cost.
-
Choose Purchase Provisioned Throughput.
-
Review the note that appears and acknowledge the commitment duration and price by selecting the checkbox. Then choose Confirm purchase.
-
The console displays the Provisioned Throughput overview page. The Status of the Provisioned Throughput in the Provisioned Throughput table becomes Creating. When the Provisioned Throughput is finished being created, the Status becomes In service. If the update fails, the Status becomes Failed.
- API
-
To purchase a Provisioned Throughput, send a CreateProvisionedModelThroughput request with an Amazon Bedrock control plane endpoint.
The following table briefly describes the parameters and request body CreateProvisionedModelThroughput request syntax):
Variable |
Required? |
Use case |
modelId |
Yes |
To specify the base model ID or ARN for purchasing Provisioned Throughput, or the custom model name
or ARN |
modelUnits |
Yes |
To specify the number of model units (MUs) to purchase. To increase the number of MUs that you can purchase, visit the AWS support center to request an increase in the number of MUs that you can purchase |
provisionedModelName |
Yes |
To specify a name for the Provisioned Throughput |
commitmentDuration |
No |
To specify the duration for which to commit to the
Provisioned Throughput. Omit this field to opt for no-commitment pricing |
tags |
No |
To associate tags with your Provisioned Throughput |
clientRequestToken |
No |
To prevent reduplication of the request |
The response returns a provisionedModelArn
that you can use as a modelId
in model inference. To check when the Provisioned Throughput is ready for use, send a GetProvisionedModelThroughput request and check that the status is InService
. If the update fails, its status will be Failed
, and the GetProvisionedModelThroughput response will contain a failureMessage
.
See code examples