---
title: "POST /subscriptions/{subscriptionId}/providers/Microsoft.CognitiveServices/calculateModelCapacity"
method: POST
path: "/subscriptions/{subscriptionId}/providers/Microsoft.CognitiveServices/calculateModelCapacity"
tags: ["CognitiveServicesAccounts"]
---

# POST /subscriptions/{subscriptionId}/providers/Microsoft.CognitiveServices/calculateModelCapacity

`POST /subscriptions/{subscriptionId}/providers/Microsoft.CognitiveServices/calculateModelCapacity`

Model capacity calculator.

## Parameters

- `../../../../../common-types/resource-management/v3/types.json#/components/parameters/ApiVersionParameter` — unresolved $ref
- `../../../../../common-types/resource-management/v3/types.json#/components/parameters/SubscriptionIdParameter` — unresolved $ref

## Request body

- CalculateModelCapacityParameter — Calculate Model Capacity parameter.
  - `model` DeploymentModel — Properties of Cognitive Services account deployment model.
    - `publisher` string — Deployment model publisher.
    - `format` string — Deployment model format.
    - `name` string — Deployment model name.
    - `version` string — Optional. Deployment model version. If version is not specified, a default version will be assigned. The default version is different for different models and might change when there is new version available for a model. Default version for a model could be found from list models API.
    - `source` string — Optional. Deployment model source ARM resource ID.
    - `sourceAccount` string, arm-id — Optional. Source of the model, another Microsoft.CognitiveServices accounts ARM resource ID.
    - `callRateLimit` CallRateLimit — The call rate limit Cognitive Services account.
      - `count` number, float — The count value of Call Rate Limit.
      - `renewalPeriod` number, float — The renewal period in seconds of Call Rate Limit.
      - `rules` ThrottlingRule[]
        - `key` string
        - `renewalPeriod` number, float
        - `count` number, float
        - `minCount` number, float
        - `dynamicThrottlingEnabled` boolean
        - `matchPatterns` RequestMatchPattern[]
          - `path` string
          - `method` string
  - `skuName` string — The name of SKU.
  - `workloads` ModelCapacityCalculatorWorkload[] — List of Model Capacity Calculator Workload.
    - `requestPerMinute` integer — Request per minute.
    - `requestParameters` ModelCapacityCalculatorWorkloadRequestParam — Dictionary, Model Capacity Calculator Workload Parameters.
      - `avgPromptTokens` integer — Average prompt tokens.
      - `avgGeneratedTokens` integer — Average generated tokens.

## Response `200`

The request has succeeded.

- CalculateModelCapacityResult — Calculate Model Capacity result.
  - `model` DeploymentModel — Properties of Cognitive Services account deployment model.
    - `publisher` string — Deployment model publisher.
    - `format` string — Deployment model format.
    - `name` string — Deployment model name.
    - `version` string — Optional. Deployment model version. If version is not specified, a default version will be assigned. The default version is different for different models and might change when there is new version available for a model. Default version for a model could be found from list models API.
    - `source` string — Optional. Deployment model source ARM resource ID.
    - `sourceAccount` string, arm-id — Optional. Source of the model, another Microsoft.CognitiveServices accounts ARM resource ID.
    - `callRateLimit` CallRateLimit — The call rate limit Cognitive Services account.
      - `count` number, float — The count value of Call Rate Limit.
      - `renewalPeriod` number, float — The renewal period in seconds of Call Rate Limit.
      - `rules` ThrottlingRule[]
        - `key` string
        - `renewalPeriod` number, float
        - `count` number, float
        - `minCount` number, float
        - `dynamicThrottlingEnabled` boolean
        - `matchPatterns` RequestMatchPattern[]
          - `path` string
          - `method` string
  - `skuName` string
  - `estimatedCapacity` CalculateModelCapacityResultEstimatedCapacity — Model Estimated Capacity.
    - `value` integer
    - `deployableValue` integer

## Other responses

- `default` — An unexpected error response.

---

[API](https://skmtc.net/azure/apis/cognitiveservices-microsoft-cognitiveservices.md) · [All operations](https://skmtc.net/azure/apis/cognitiveservices-microsoft-cognitiveservices/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/azure/cognitiveservices-microsoft-cognitiveservices/revisions/691c4877e0b8/schema)