HTTP API
ModelDescriptor
A model’s capabilities and limits.
Shape: object
| Property | Required | Type | Description | Rules |
|---|---|---|---|---|
name |
yes | string |
The model name. ModelBinding.Model matches this value. |
— |
displayName |
no | string |
The name shown in the UI. | — |
contextWindowTokens |
no | integer (int32) |
The context window’s token capacity. | pattern `^-?(?:0\ |
maxOutputTokens |
no | integer (int32) |
The maximum number of tokens producible in a single response. | pattern `^-?(?:0\ |
supportsStreaming |
no | boolean |
Whether streaming responses are supported. | — |
supportsTools |
no | boolean |
Whether tool calling is supported. | — |
supportsReasoning |
no | boolean |
Whether the reasoning-effort setting is supported. | — |
supportsStructuredOutput |
no | boolean |
Whether output conforming to a JSON schema can be produced. | — |
supportsSampling |
no | boolean |
Whether the model accepts the sampling parameters temperature and topP; null when the catalog does not say. |
— |
inputCostPerMillionTokens |
no | number (double) |
The cost per million input tokens. For reporting only. | pattern `^-?(?:0\ |
outputCostPerMillionTokens |
no | number (double) |
The cost per million output tokens. For reporting only. | pattern `^-?(?:0\ |
cachedInputCostPerMillionTokens |
no | number (double) |
The cost per million input tokens that were served from the provider’s prompt cache. For reporting only. | pattern `^-?(?:0\ |