Request payload for the Responses API.
| Property | Type | Description |
|---|---|---|
| input | A list of input items. At least one of input and instructions is required. | |
| instructions | String | A system-level instruction, inserted as the first system message of the model context. |
| max_output_tokens | Integer | An upper bound for the number of tokens that can be generated, including visible output and reasoning tokens. |
| model | String | ID of the model to use (e.g., deepseek-flash, deepseek-v4-pro). |
| reasoning | Response_Reasoning | Configuration of the thinking mode. |
| stream | Boolean | If set, the response is streamed as semantic server-sent events. |
| temperature | Number | Sampling temperature between 0 and 2. |
| text | Response_Text | Configuration of the text output. |
| tool_choice | Any | Controls which (if any) tool is called by the model. Either the string none, auto or required, or a Response_Tool_Choice object. |
| tools | A list of tools the model may call. | |
| top_logprobs | Integer | An integer between 0 and 20 specifying the number of most likely tokens to return at each token position. |
| top_p | Number | Nucleus sampling probability mass (between 0 and 1). |
| user | String | A custom end-user identifier ([a-zA-Z0-9-_], max 512 chars). |