Module request 

Request body and POST method for the Responses API (POST /responses).

The request format follows the OpenAI Responses API as implemented by the official Python SDK, restricted to the parameters documented by OpenAI and the compatible providers (DeepSeek, Qwen). Provider-proprietary parameters are opt-in via the deepseek / qwen cargo features.

Structs

ReasoningConfig

Configuration options for reasoning models.

ReasoningTextPart

A plaintext reasoning content part.

RequestBody

Creates a model response for the given input.

SummaryTextPart

A reasoning summary content part.

TextConfig

Configuration options for a text response from the model.

Enums

FunctionCallOutputContent

The output content of a function (or custom tool) call: either plain text or an array of content parts.

ImageDetail

Controls how an input image is processed. low downscales the image before inference (faster, fewer tokens); the other values keep the original image.

ImageUrl

The source of an input image. DeepSeek and Qwen take a plain URL string; OpenAI takes an object with a url field.

Input

The input parameter: either a plain string (treated as a single user message) or a list of input items.

InputContentPart

A content part of an input message or tool output.

InputItem

An item of the input list.

MessageContent

The content of an input message: either plain text or an array of content parts.

Role

The role of a message author.

SummaryEffort

The verbosity of reasoning summaries.

TextFormat

The format of the text output.

Tool

A tool the model may call while generating a response.

ToolChoice

How the model should select which tool (or tools) to use.

ToolChoiceMode

The tool choice mode.

ToolChoiceSpecific

A specific tool the model is forced to call.

Truncation

The truncation strategy to use when the input exceeds the model’s context window.