# Training plans creation To reserve compute capacity using the SageMaker training plans capability, follow these steps: 1. **Identify your target resource: ** Begin by determining whether you need capacity for SageMaker training jobs or SageMaker HyperPod clusters. 1. **Specify your capacity requirements :** Define your capacity needs in detail. This includes selecting the appropriate instance type for your workload, determining the number of instances required, and specifying the duration of use. For information about the supported instance types in a given AWS Region, duration, and quantity options, see [Supported instance types, AWS Regions, and pricing](reserve-capacity-with-training-plans.md#training-plans-supported-instances-and-regions). 1. **Search for available training plan offerings: ** Once you specify your requirements, use SageMaker training plans' search functionality to find available training plan offerings across one or more segments. Each offering includes details such as start time, specific availability zone for Reserved Capacity, and plan price. Review these offerings, considering factors like cost-effectiveness, geographical preferences, and alignment with your specified needs. If no suitable plan is available, adjust your search criteria and look for a new set of offerings. 1. **Create a training plan based on a suitable offering: ** After identifying a suitable offering, proceed to create your training plan. This process involves selecting your chosen offering and initiating the reservation. + The training plan reservation creates an invoice. + The payment for the total amount is collected as part of the fulfillment process. Once the payment is completed, the plan is ready for scheduling your SageMaker training jobs or creating HyperPod clusters. To learn about how to use training plans for your SageMaker training jobs , see [Training plans utilization for SageMaker training jobs](training-plan-utilization-for-training-jobs.md). To learn about how to use training plans for your HyperPod clusters, see [Training plans utilization for Amazon SageMaker HyperPod clusters](training-plan-utilization-for-hyperpod.md). You can create a training plan using either the SageMaker AI console or programmatic methods. The SageMaker AI console offers a visual, graphical interface with a comprehensive view of your options, while programmatic creation can be done using the AWS CLI or SageMaker SDKs to interact directly with the training plans API. For step-by-step console instructions and detailed API references, refer to the respective sections in this documentation. **Topics** + [SageMaker training plans creation using the SageMaker AI console](training-plan-creation-using-console.md) + [SageMaker training plans creation using the SageMaker API, or AWS CLI](training-plan-creation-using-api-cli-sdk.md) # SageMaker training plans creation using the SageMaker AI console SageMaker training plans offer a convenient way to create training plans through the SageMaker AI console UI, allowing users to easily schedule their machine learning training resources. This guide walks you through the process of creating a training plan for SageMaker training jobs and SageMaker HyperPod clusters using the SageMaker AI console. By following these steps, you will search for training plan offerings, review available options, and purchase the plan that best fits your needs. To create a training plan visually using a UI: 1. Start by navigating to the SageMaker AI console at [https://console.aws.amazon.com/sagemaker/](https://console.aws.amazon.com/sagemaker/). 1. Choose **Training Plans** in the left pane menu. 1. From there, choose the **Create training plan** button in the main content area to start the process of setting up your customized training schedule. ![\[SageMaker AI console showing the training plans page. The interface displays information about how training plans work, including steps to request, monitor, and use a plan. The left navigation pane highlights the "Training Plans" option, and a "Create training plan" button is visible in the main content area.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/training-plans/tp-console.png) Next, search for plan offerings that match your compute requirements. **Topics** + [Search training plan offerings](search-training-plan-offerings.md) + [Reserve the best training plan](choose-best-training-plan.md) + [List training plans](list-training-plans.md) + [View training plan details](training-plan-details.md) # Search training plan offerings After you choose **Training Plans** in the left pane of the SageMaker AI console, and then **Create training plan**, a **Find training plan** form opens up. This form allows you to specify your requirements and search for suitable training plan offerings. Follow these steps to complete the form: 1. Identify your **Target**: Training plans are specific to their target resource. Specify whether you want to use a plan to run SageMaker training jobs or SageMaker HyperPod clusters. 1. For **Compute type**, you can choose between **Instance** or **UltraServer**. UltraServers are connect multiple Amazon EC2 instances using a low-latency, high-bandwidth accelerator interconnect. For more information, see [Amazon EC2 UltraServers](https://aws.amazon.com/ec2/ultraservers/). To learn about how you can use UltraServers with SageMaker AI, see [UltraServers in SageMaker AI](reserve-capacity-with-training-plans.md#training-plans-ultraservers). 1. Choose your preferred **Instance type** and **Instance count**: For available instance types in a given AWS Region, duration, and quantity options, see [Supported instance types, AWS Regions, and pricing](reserve-capacity-with-training-plans.md#training-plans-supported-instances-and-regions). 1. Define your time parameters: Choose your desired start and end dates, and specify the plan duration within this window. 1. Choose **Find training plans**. ![\[SageMaker AI console showing the Search training plan offerings page. The interface displays options to select the target resource for the plan (Training job or HyperPod cluster), specify instance type and count, set start and end dates, and input duration. A Find training plans button is visible at the bottom of the form.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/training-plans/tp-search-training-plan-offerings.png) SageMaker training plans search for offerings that match your capacity requirements. When matches are found within your specified time frame, they appear at the bottom of the page. Each training plan offering includes the following details: + Total plan duration + Start and end dates + Total upfront price: Hover over the price to view the detailed breakdown of instance hourly rate, instance count, and total hours + Total number of plan segments Clicking the segment detail link opens a modal view with segment-specific details: + Duration + Start and end dates + Availability zone ![\[SageMaker AI console showing the Search training plan offerings page with input fields for plan requirements and the Available plans section displaying details of three found plans with varying durations, prices, and availability status.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/training-plans/tp-available-offerings.png) If no suitable plans are found or the available plans don't meet your needs, adjust your search criteria by modifying the parameters in the **Training plans requirements** form. Once you find a suitable offering, select it and choose **Next** to continue to the plan reservation page. On this page, you can name your plan, and then review and confirm your selection before finalizing your reservation. **Note** Plans marked `Immediately available` will start within 30 minutes, provided payment is completed no less than 5 minutes before the scheduled start time. # Reserve the best training plan The search of a training plan has returned offerings that fit your capacity needs and budget. 1. Enter a name for your plan and then choose **Next**. 1. Review and **Submit** your purchase order. **Important** Training plans cannot be modified once purchased. Training plans cannot be shared across AWS accounts or within your AWS Organization. After submitting your order + The training plan initially appears as `Pending` in your training plan list. + An invoice is generated automatically upon order receipt. + The total payment is collected during the fulfillment process. + Once payment is successfully processed, the plan status changes to `Scheduled` and the plan becomes available for use. ![\[SageMaker AI console showing the "Review and purchase" page for a training plan. The page displays training plan details, segment information, price, plan name, and tags. Options to edit, cancel, go back, or create the plan are available.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/training-plans/tp-review-and-purchase-training-plan.png) # List training plans To view your training plans: 1. Navigate to the SageMaker AI console at [https://console.aws.amazon.com/sagemaker/](https://console.aws.amazon.com/sagemaker/). 1. Choose **Training Plans** in the left pane menu. This displays a list of all your training plans, including their names, status, target resource type, and other key details. After purchasing a plan, you are directed to this list. Newly created plans appear with a `Pending` status until payment is completed. The status is typically updated within a few minutes of payment processing. ![\[SageMaker AI console showing the list of training plans page. The page includes a table listing the training plans with details such as name, status, total instances, in-use instances, zone, start date, and end date. A button to create a new training plan is visible.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/training-plans/tp-list-training-plans.png) # View training plan details From the training plans list, follow a plan's name to view its details. Specifically, you can check your current capacity usage, and list your workloads in your plan's details page. The details page shows: + The training plan overview: Status, target, instance type, and duration. + Expandable sections for segment details, pricing, plan name, and tags. + Capacity utilization: + Total: The total number of instances reserved in this training plan. + In-use: The number of instances currently in use from this training plan. + Available instances: The number of instances currently available for use in this training plan. At the bottom of the page, a link allows you to view either the training jobs or the list of SageMaker HyperPod cluster instance groups associated with this plan, depending on its target resource. ![\[SageMaker AI console page displaying details of a training plan. The page shows basic plan information, status, and instance details. Below are expandable sections for additional details. At the bottom, a capacity utilization section shows total, in-use, and available instances for the plan.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/training-plans/tp-view-training-plan.png) # SageMaker training plans creation using the SageMaker API, or AWS CLI SageMaker training plans support the programmatic creation of training plans through its API. You can interact with the training plans API using the AWS CLI or SageMaker SDKs. SageMaker training plans's API actions provide a comprehensive workflow for managing training plans programmatically: + **`SearchTrainingPlanOfferings`:** Enables users to query and discover available compute resources by specifying parameters like instance type, count, and desired time window. The API returns a ranked list of training plan offerings that best match the user's requirements. + **`CreateTrainingPlan`:** Allows reservation of a specific training plan offering, transforming a potential compute capacity into scheduled reserved capacities with a unique training plan ARN. + **`ListTrainingPlans`:** Provides a method to retrieve and review all existing training plans in a user's AWS account, with optional filtering and sorting capabilities. + **`DescribeTrainingPlan`:** Offers detailed insights into a specific training plan, including its lifecycle stages from `Pending` to `Active` to `Expired`. + **`ExtendTrainingPlan`:** Extends an existing training plan by purchasing an extension offering. For more information, see [Training plans extension](training-plan-extension.md). + **`DescribeTrainingPlanExtensionHistory`:** Retrieves the extension history for a training plan. For more information, see [Training plans extension](training-plan-extension.md). **Topics** + [Search training plan offerings](search-training-plan-offerings-api-cli-sdk.md) + [Reserve the best training plan](choose-best-training-plan-using-api-cli-sdk.md) + [List training plans](list-training-plans-using-api-cli-sdk.md) + [View training plan details](training-plan-details-using-api-cli-sdk.md) # Search training plan offerings To create a training plan, start by calling the [https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_SearchTrainingPlanOfferings.html](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_SearchTrainingPlanOfferings.html) API operation, passing your plan requirements (such as instance type, count, and desired time window) as input parameters. Training plans are specific to their target resource. Ensure that you specify which target resource the plan will be used for (`training-job` or `hyperpod-cluster`). The API returns a list of available offerings that match your requirements. If no suitable offerings are found, you may need to adjust your requirements and search again. This API call retrieves the training plan offerings that best meet your capacity needs. Each [https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_TrainingPlanOffering.html](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_TrainingPlanOffering.html) returned in the response is identified by a unique offering ID. The first offering in the list represents the best match for your requirements. If no suitable training plan is available within your specified dates, the list is empty. Adjust your search criteria and look for a new set of offerings. + Reservation durations are available in 1-day increments from 1 to 182 days. + The reservation instance quantity options are 1, 2, 4, 8, 16, 32 or 64 instances. To learn about the list of available instances supported by SageMaker training plans, see [Supported instance types, AWS Regions, and pricing](reserve-capacity-with-training-plans.md#training-plans-supported-instances-and-regions). The following example uses an AWS CLI command to request training plan offerings with a specified instance type, count, and time information. ``` # List training plan offerings with instance type, instance count, duration in hours, start time after, and end time before. aws sagemaker search-training-plan-offerings \ --target-resources "training-job" \ --instance-type "ml.p4d.24xlarge" \ --instance-count 1 \ --duration-hours 15 \ --start-time-after "1737484800" --end-time-before "1737657600" ``` This JSON document is a sample response from the SageMaker training plans API. The response provides information about multiple available training plan offerings that match the specified capacity requirements. It includes three distinct offerings with varying durations, upfront fees, and start/end times, all using the same instance type and targeting training jobs. ``` { "TrainingPlanOfferings": [ { "TrainingPlanOfferingId": "tpo-SHA-256-hash-value", "TargetResources": [ "training-job" ], "RequestedStartTimeAfter": "2025-01-21T11:08:27.704000-08:00", "DurationHours": 15, "DurationMinutes": 51, "UpfrontFee": "xxxx.xx", "CurrencyCode": "USD", "ReservedCapacityOfferings": [ { "InstanceType": "ml.p4d.24xlarge", "InstanceCount": 1, "AvailabilityZone": "us-west-2a", "DurationHours": 15, "DurationMinutes": 51, "StartTime": "2025-01-21T11:39:00-08:00", "EndTime": "2025-01-22T03:30:00-08:00" } ] }, { "TrainingPlanOfferingId": "tpo-SHA-256-hash-value", "TargetResources": [ "training-job" ], "RequestedStartTimeAfter": "2025-01-21T11:08:27.704000-08:00", "DurationHours": 39, "DurationMinutes": 51, "UpfrontFee": "xxxx.xx", "CurrencyCode": "USD", "ReservedCapacityOfferings": [ { "InstanceType": "ml.p4d.24xlarge", "InstanceCount": 1, "AvailabilityZone": "us-west-2a", "DurationHours": 39, "DurationMinutes": 51, "StartTime": "2025-01-21T11:39:00-08:00", "EndTime": "2025-01-23T03:30:00-08:00" } ] }, { "TrainingPlanOfferingId": "tpo-SHA-256-hash-value", "TargetResources": [ "training-job" ], "RequestedStartTimeAfter": "2025-01-21T11:08:27.704000-08:00", "DurationHours": 24, "DurationMinutes": 0, "UpfrontFee": "xxxx.xx", "CurrencyCode": "USD", "ReservedCapacityOfferings": [ { "InstanceType": "ml.p4d.24xlarge", "InstanceCount": 1, "AvailabilityZone": "us-west-2a", "DurationHours": 24, "DurationMinutes": 0, "StartTime": "2025-01-22T03:30:00-08:00", "EndTime": "2025-01-23T03:30:00-08:00" } ] } ] } ``` The following is a sample command of how to use the AWS CLI to search for training plan offerings that include UltraServers. ``` aws sagemaker search-training-plan-offerings \ --ultra-server-type ml.c6i-32xlargesc \ --ultra-server-count 1 \ --duration-hours 24 \ --target-resources hyperpod-cluster --start-time-after "1737484800" \ --end-time-before "1737657600" ``` ``` { "TrainingPlanOfferings": [ { "TrainingPlanOfferingId": "tpo-SHA-256-hash-value", "TargetResources": [ "training-job" ], "RequestedStartTimeAfter": "2025-07-21T16:59:25.760000+00:00", "DurationHours": 24, "DurationMinutes": 0, "UpfrontFee": "0.24", "CurrencyCode": "USD", "ReservedCapacityOfferings": [ { "ReservedCapacityType": "UltraServer", "UltraServerType": "ml.u-p6e-gb200x72", "UltraServerCount": 1, "InstanceType": "ml.p6e-gb200.36xlarge", "InstanceCount": 18, "AvailabilityZone": "us-east-2a", "DurationHours": 24, "DurationMinutes": 0, "StartTime": "2025-07-22T11:30:00+00:00", "EndTime": "2025-07-23T11:30:00+00:00" } ] } ] } ``` The following sections define the mandatory and optional input request parameters for the `SearchTrainingPlanOfferings` API operation. ## Required parameters When calling the [https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_SearchTrainingPlanOfferings.html](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_SearchTrainingPlanOfferings.html) API to list training plan offerings that meet your requirements, you must provide the following values: + `TargetResources`: The target resources (`training-job` or `hyperpod-cluster`) for which the plan will be used. The default value is `training-job`. Training plans are specific to their target resource. + A training plan designed for SageMaker training jobs can only be used to schedule and run training jobs. + A training plan for HyperPod clusters can be used exclusively to provide compute resources to a cluster's instance group. + `InstanceType`: The type of instance to provision. The `InstanceType` must be of a supported type. To learn about the list of available instances supported by SageMaker training plans, see [Supported instance types, AWS Regions, and pricing](reserve-capacity-with-training-plans.md#training-plans-supported-instances-and-regions). + `InstanceCount`: The number of instances to provision. If the number of instances is greater than 1, it should be a power of 2. + `DurationHour`: The total duration of your requested plan in hours. The `DurationHour` is rounded up to the nearest multiple of 24. ## Optional parameters The following sections provide details on some optional parameters that you can pass to your `SearchTrainingPlanOfferings` API request. + `StartTimeAfter`: Specify the requested start time of the plan. The `StartTimeAfter` should be a `timestamp` or an `ISO 8601 date/time` value in the future. + `EndTimeBefore`: Specify the requested end time of the plan in a `timestamp` or an `ISO 8601 date/time` format. The `EndTimeBefore` should be at least 24 hours after the start time . + `UltraServerType` : Specify the type of UltraServer to search for. For more information about UltraServers, see [UltraServers in SageMaker AI](reserve-capacity-with-training-plans.md#training-plans-ultraservers). + `UltraServerCount`: Specify the number of UltraServers to search for. # Reserve the best training plan After reviewing the available training plan offerings that best match your requirements, you can reserve a specific plan by calling the [https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateTrainingPlan.html](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateTrainingPlan.html) API operation. When created, the plan initially enters a `Pending` state and remains there until the reservation process is complete. The response to the API call returns a training plan Amazon Resource Name (ARN). Make a note of this ARN for tracking and monitoring purposes later on. The training plan reservation is fulfilled asynchronously in the backend. The payment for the total amount is automatically collected as part of the fulfillment process. Once the payment transaction is completed and the requested reserved capacities are secured, the training plan is set to the `Scheduled` state, and is ready for scheduling. **Important** Training plans cannot be modified once purchased. Training plans cannot be shared across AWS accounts or within your AWS Organization. The following example uses the an AWS CLI command to request a specific training plan, passing the plan ID as a parameter. ``` aws sagemaker create-training-plan \ --training-plan-offering-id "tpo-SHA-256-hash-value" \ --training-plan-name "name" \ ``` This JSON document is a sample response from the SageMaker training plans API. The response contains the Amazon Resource Name (ARN) of the training plan that has been successfully created. **Note** The training plan remains in a `Pending` status until the fulfillment process is complete. ``` { "TrainingPlanArn":"arn:aws:sagemaker:us-east-1:123456789123:training-plan/large-models-fine-tuning" } ``` The following sections define the mandatory and optional input request parameters for the [https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateTrainingPlan.html](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateTrainingPlan.html) API operation. ## Required parameters When calling [https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateTrainingPlan.html](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateTrainingPlan.html) API to reserve a particular training plan, you must provide the following values: + `TrainingPlanOfferingId`: The ID of the plan you are choosing. You can retrieve the ID of a plan offering in the response of your `SearchTrainingPlanOfferings` API call. Its format should start with `pto-*`. + `TrainingPlanName`: The name of the plan you are creating. # List training plans You can list all the training plans that have been created in your AWS account and Region by calling the [https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_ListTrainingPlans.html](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_ListTrainingPlans.html) API. The following example uses an AWS CLI command to retrieve the list of your training plans. ``` aws sagemaker list-training-plans \ --start-time-after "2024-09-26T00:00:01.000Z" ``` This JSON document is a sample response from the SageMaker training plans API. The response provides details about one training plan that has been successfully created and reserved. ``` { "[TrainingPlanSummaries](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_TrainingPlanSummary.html)": [ { "AvailableInstanceCount": 2, "CurrencyCode": "USD", "DurationHours": 48, "DurationMinutes": 0, "EndTime": "2024-09-28T04:30:00-07:00", "InUseInstanceCount": 2, "[ReservedCapacitySummaries](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_ReservedCapacitySummary.html)": [ { "AvailabilityZone": "string", "DurationHours": 48, "DurationMinutes": 0, "EndTime": "2024-09-28T04:30:00-07:00", "InstanceType": "ml.p5.48xlarge", "ReservedCapacityArn": "arn:aws:sagemaker:us-east-1:123456789123:reserved-capacity/large-models-fine-tuning-rc1", "StartTime": "2024-09-26T04:30:00-07:00", "Status": "Scheduled", "TotalInstanceCount": 4, "UltraServerCount": 4, "UltraServerType": "ml.p6e-gb200.36xlarge" } ], "StartTime": "2024-09-26T04:30:00-07:00", "Status": "Scheduled", "StatusMessage": "Payment confirmed, training plan scheduled." "TargetResources": [ "training-job" ], "TotalInstanceCount": 4, "TotalUltraServerCount": 4, "TrainingPlanArn": "arn:aws:sagemaker:us-east-1:123456789123:training-plan/large-models-fine-tuning", "TrainingPlanName": "large-models-fine-tuning", "UpfrontFee": "xxxx.xx" } ] } ``` The following sections provide details of some optional parameters that you can pass to your `ListTrainingPlans` API request. ## Optional parameters The following sections provide details on some optional parameters that you can pass to your `ListTrainingPlans` API request. + `StartTimeAfter`: The start time of the actual time range of the listed plans, specified as a `timestamp` or an `ISO 8601 date/time`. + `StartTimeBefore`: The end time of the actual time range of the listed plans, specified as a `timestamp` or an `ISO 8601 date/time`. + `Filters`: Criteria used to filter the results, with up to 5 Name-Value pairs where "Name" is the name of a field of a [TrainingPlanSummary](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_TrainingPlanSummary.html) and "Value" is the value to consider for the filter. For example `Name=Status,Value=Active`. The following example uses an AWS CLI command to retrieve your list of training plans, using some of the optional parameters described above. ``` aws sagemaker list-training-plans --max-results 10 --sort-by StartTime --sort-order Descending --start-time-after 13000000 --filters Name=Status,Value=Active ``` # View training plan details To monitor the status or retrieve details of a training plan, you can use the [https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeTrainingPlan.html](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeTrainingPlan.html) API. The API response includes a `Status` field, which reflects the current state of the training plan: + If the plan purchase fails, the status is set to `Failed`. + Upon successful payment, the status transitions from `Pending` to `Scheduled`, based on the plan's start date. + When the plan reaches its start date, the status changes to `Active`. + For plans with multiple discontinuous reserved capacities, the status reverts to `Scheduled` between active periods, until the start date of the next reserved capacity. + After the plan's end date, the status becomes `Expired`. Once the status is `Scheduled`, you can utilize the capacity reserved in the plan for your SageMaker training jobs or HyperPod cluster workloads. **Note** Training jobs associated with the plan remain in `Pending` status until the plan becomes `Active`. For HyperPod clusters using a training plan for compute capacity, the instance group status appears as `InService` once created. The following example uses an AWS CLI command to retrieve the details of a training plan by its name. ``` aws sagemaker describe-training-plan \ --training-plan-name "name" ``` This JSON document is a sample response from the SageMaker training plans API. This response provides details about a training plan that has been successfully created. ``` { "AvailableInstanceCount": 2, "CurrencyCode": "USD", "DurationHours": 48, "DurationMinutes": 0, "EndTime": "2024-09-28T04:30:00-07:00", "InUseInstanceCount": 2, "[ReservedCapacitySummaries](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_ReservedCapacitySummary.html)": [ { "AvailabilityZone": "string", "DurationHours": 48, "DurationMinutes": 0, "EndTime": "2024-09-28T04:30:00-07:00", "InstanceType": "ml.p5.48xlarge", "ReservedCapacityArn": "arn:aws:sagemaker:us-east-1:123456789123:reserved-capacity/large-models-fine-tuning-rc1", "StartTime": "2024-09-26T04:30:00-07:00", "Status": "Scheduled", "TotalInstanceCount": 4, "UltraServerCount": 4, "UltraServerType": "ml.p6e-gb200.36xlarge" } ], "StartTime": "2024-09-26T04:30:00-07:00", "Status": "Scheduled", "StatusMessage": "Payment confirmed, training plan scheduled." "TargetResources": [ "training-job" ], "TotalInstanceCount": 4, "TotalUltraServerCount": 4, "TrainingPlanArn": "arn:aws:sagemaker:us-east-1:123456789123:training-plan/large-models-fine-tuning", "TrainingPlanName": "large-models-fine-tuning", "UpfrontFee": "xxxx.xx" } ``` The following sections define the mandatory input request parameter for the `DescribeTrainingPlan` API operation. ## Required parameters + `TrainingPlanName`: The name of the training plan you want to describe.