235 235
236* `context_management` (array | null) — Optional context-management directives (e.g. compaction). Parsed but not yet executed.236* `context_management` (array | null) — Optional context-management directives (e.g. compaction). Parsed but not yet executed.
237 237
238* `include` (array | null) — What additional output data to include in the response. Currently the only supported value is \`reasoning.encrypted\_content\` which returns an encrypted version of the reasoning tokens.238* `include` (array | null) — What additional output data to include in the response. Supported values include
239 \`reasoning.encrypted\_content\` (encrypted reasoning tokens) and tool-output options.
240 OpenAI's \`message.output\_text.logprobs\` is accepted for compatibility but silently ignored.
239 241
240* `input` (string | array\<object | object | object | object | object>, required) — Content of the input passed to a \`/v1/response\` request.242* `input` (string | array\<object | object | object | object | object>, required) — Content of the input passed to a \`/v1/response\` request.
241 243
344 346
345* `object` (string, required) — The object type of this resource. Always set to \`response\`.347* `object` (string, required) — The object type of this resource. Always set to \`response\`.
346 348
347* `output` (array\<object | object | object | object | object | object | object | object | object>, required) — The response generated by the model.349* `output` (array\<object | object | object | object | object | object | object | object | object | object>, required) — The response generated by the model.
348 350
349* `parallel_tool_calls` (boolean, required) — Whether to allow the model to run parallel tool calls.351* `parallel_tool_calls` (boolean, required) — Whether to allow the model to run parallel tool calls.
350 352
382 384
383 * `type` (string, required) — Type is always \`"function"\`.385 * `type` (string, required) — Type is always \`"function"\`.
384 386
385* `tools` (array\<object | object | object | object | object | object | object>, required) — A list of tools the model may call in JSON-schema. Currently, only functions and web search are supported as tools. A max of 128 tools are supported.387* `tools` (array\<object | object | object | object | object | object | object | object>, required) — A list of tools the model may call in JSON-schema. Currently, only functions and web search are supported as tools. A max of 128 tools are supported.
386 388
387* `top_logprobs` (integer, required) — An integer between 0 and 8 specifying the number of most likely tokens to return at each token position.389* `top_logprobs` (integer, required) — An integer between 0 and 8 specifying the number of most likely tokens to return at each token position.
388 390
430 432
431 * `file_search_calls` (integer, required) — Number of file search calls.433 * `file_search_calls` (integer, required) — Number of file search calls.
432 434
435 * `image_generation_calls` (integer, required) — Number of image generation calls.
436
433 * `mcp_calls` (integer, required) — Number of MCP calls.437 * `mcp_calls` (integer, required) — Number of MCP calls.
434 438
435 * `web_search_calls` (integer, required) — Number of web search calls.439 * `web_search_calls` (integer, required) — Number of web search calls.
730 734
731* `object` (string, required) — The object type of this resource. Always set to \`response\`.735* `object` (string, required) — The object type of this resource. Always set to \`response\`.
732 736
733* `output` (array\<object | object | object | object | object | object | object | object | object>, required) — The response generated by the model.737* `output` (array\<object | object | object | object | object | object | object | object | object | object>, required) — The response generated by the model.
734 738
735* `parallel_tool_calls` (boolean, required) — Whether to allow the model to run parallel tool calls.739* `parallel_tool_calls` (boolean, required) — Whether to allow the model to run parallel tool calls.
736 740
768 772
769 * `type` (string, required) — Type is always \`"function"\`.773 * `type` (string, required) — Type is always \`"function"\`.
770 774
771* `tools` (array\<object | object | object | object | object | object | object>, required) — A list of tools the model may call in JSON-schema. Currently, only functions and web search are supported as tools. A max of 128 tools are supported.775* `tools` (array\<object | object | object | object | object | object | object | object>, required) — A list of tools the model may call in JSON-schema. Currently, only functions and web search are supported as tools. A max of 128 tools are supported.
772 776
773* `top_logprobs` (integer, required) — An integer between 0 and 8 specifying the number of most likely tokens to return at each token position.777* `top_logprobs` (integer, required) — An integer between 0 and 8 specifying the number of most likely tokens to return at each token position.
774 778
816 820
817 * `file_search_calls` (integer, required) — Number of file search calls.821 * `file_search_calls` (integer, required) — Number of file search calls.
818 822
823 * `image_generation_calls` (integer, required) — Number of image generation calls.
824
819 * `mcp_calls` (integer, required) — Number of MCP calls.825 * `mcp_calls` (integer, required) — Number of MCP calls.
820 826
821 * `web_search_calls` (integer, required) — Number of web search calls.827 * `web_search_calls` (integer, required) — Number of web search calls.