Reference 2026-09-08 01:58 UTC to 2026-09-09 16:02 UTC

442 442 

443 - `parameters: optional FunctionParameters`443 - `parameters: optional FunctionParameters`

444 444 

445 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.445 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/api/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

446 446 

447 Omitting `parameters` defines a function with an empty parameter list.447 Omitting `parameters` defines a function with an empty parameter list.

448 448 

449 - `strict: optional boolean or null`449 - `strict: optional boolean or null`

450 450 

451 Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the `parameters` field. Only a subset of JSON Schema is supported when `strict` is `true`. Learn more about Structured Outputs in the [function calling guide](/docs/guides/function-calling).451 Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the `parameters` field. Only a subset of JSON Schema is supported when `strict` is `true`. Learn more about Structured Outputs in the [function calling guide](/api/docs/guides/function-calling).

452 452 

453### Function Parameters453### Function Parameters

454 454 

455- `FunctionParameters = map[unknown]`455- `FunctionParameters = map[unknown]`

456 456 

457 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.457 The parameters the functions accepts, described as a JSON Schema object. See the [guide](/api/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

458 458 

459 Omitting `parameters` defines a function with an empty parameter list.459 Omitting `parameters` defines a function with an empty parameter list.

460 460 

474- `Reasoning object { context, effort, generate_summary, 2 more }`474- `Reasoning object { context, effort, generate_summary, 2 more }`

475 475 

476 Configuration options for476 Configuration options for

477 [reasoning models](https://platform.openai.com/docs/guides/reasoning).477 [reasoning models](/api/docs/guides/reasoning).

478 478 

479 - `context: optional "auto" or "current_turn" or "all_turns" or null`479 - `context: optional "auto" or "current_turn" or "all_turns" or null`

480 480 

499 Reducing reasoning effort can result in faster responses and fewer tokens499 Reducing reasoning effort can result in faster responses and fewer tokens

500 used on reasoning in a response. Not all reasoning models support every500 used on reasoning in a response. Not all reasoning models support every

501 value. See the501 value. See the

502 [reasoning guide](https://platform.openai.com/docs/guides/reasoning)502 [reasoning guide](/api/docs/guides/reasoning)

503 for model-specific support.503 for model-specific support.

504 504 

505 - `"none"`505 - `"none"`

571 Reducing reasoning effort can result in faster responses and fewer tokens571 Reducing reasoning effort can result in faster responses and fewer tokens

572 used on reasoning in a response. Not all reasoning models support every572 used on reasoning in a response. Not all reasoning models support every

573 value. See the573 value. See the

574 [reasoning guide](https://platform.openai.com/docs/guides/reasoning)574 [reasoning guide](/api/docs/guides/reasoning)

575 for model-specific support.575 for model-specific support.

576 576 

577 - `"none"`577 - `"none"`

608- `ResponseFormatJSONSchema object { json_schema, type }`608- `ResponseFormatJSONSchema object { json_schema, type }`

609 609 

610 JSON Schema response format. Used to generate structured JSON responses.610 JSON Schema response format. Used to generate structured JSON responses.

611 Learn more about [Structured Outputs](/docs/guides/structured-outputs).611 Learn more about [Structured Outputs](/api/docs/guides/structured-outputs).

612 612 

613 - `json_schema: object { name, description, schema, strict }`613 - `json_schema: object { name, description, schema, strict }`

614 614 

635 If set to true, the model will always follow the exact schema defined635 If set to true, the model will always follow the exact schema defined

636 in the `schema` field. Only a subset of JSON Schema is supported when636 in the `schema` field. Only a subset of JSON Schema is supported when

637 `strict` is `true`. To learn more, read the [Structured Outputs637 `strict` is `true`. To learn more, read the [Structured Outputs

638 guide](/docs/guides/structured-outputs).638 guide](/api/docs/guides/structured-outputs).

639 639 

640 - `type: "json_schema"`640 - `type: "json_schema"`

641 641 

660- `ResponseFormatTextGrammar object { grammar, type }`660- `ResponseFormatTextGrammar object { grammar, type }`

661 661 

662 A custom grammar for the model to follow when generating text.662 A custom grammar for the model to follow when generating text.

663 Learn more in the [custom grammars guide](/docs/guides/custom-grammars).663 Learn more in the [custom grammars guide](/api/docs/guides/function-calling#context-free-grammars).

664 664 

665 - `grammar: string`665 - `grammar: string`

666 666 

677- `ResponseFormatTextPython object { type }`677- `ResponseFormatTextPython object { type }`

678 678 

679 Configure the model to generate valid Python code. See the679 Configure the model to generate valid Python code. See the

680 [custom grammars guide](/docs/guides/custom-grammars) for more details.680 [custom grammars guide](/api/docs/guides/function-calling#context-free-grammars) for more details.

681 681 

682 - `type: "python"`682 - `type: "python"`

683 683 

44 44 

45- `event_types: optional array of "api_key.created" or "api_key.updated" or "api_key.deleted" or 144 more`45- `event_types: optional array of "api_key.created" or "api_key.updated" or "api_key.deleted" or 144 more`

46 46 

47 Return only events with a `type` in one of these values. For example, `project.created`. For all options, see the documentation for the [audit log object](/docs/api-reference/audit-logs/object).47 Return only events with a `type` in one of these values. For example, `project.created`. For all options, see the documentation for the [audit log object](/api/reference/resources/admin/subresources/organization/subresources/audit_logs).

48 48 

49 - `"api_key.created"`49 - `"api_key.created"`

50 50 

16 16 

17- `geography: optional string or null`17- `geography: optional string or null`

18 18 

19 Create the project with the specified data residency region. Your organization must have access to Data residency functionality in order to use. See [data residency controls](/docs/guides/your-data#data-residency-controls) to review the functionality and limitations of setting this field.19 Create the project with the specified data residency region. Your organization must have access to Data residency functionality in order to use. See [data residency controls](/api/docs/guides/your-data#data-residency-controls) to review the functionality and limitations of setting this field.

20 Deprecated: use `residency` instead. Do not provide both `geography` and `residency`.20 Deprecated: use `residency` instead. Do not provide both `geography` and `residency`.

21 21 

22- `residency: optional ProjectResidency or null`22- `residency: optional ProjectResidency or null`

23 23 

24 Create the project with the specified residency configuration. Your organization must have access to the requested residency configuration in order to use it. See [data residency controls](/docs/guides/your-data#data-residency-controls) to review the functionality and limitations of setting this field.24 Create the project with the specified residency configuration. Your organization must have access to the requested residency configuration in order to use it. See [data residency controls](/api/docs/guides/your-data#data-residency-controls) to review the functionality and limitations of setting this field.

25 25 

26 - `"GLOBAL"`26 - `"GLOBAL"`

27 27 

120 120 

121 The redacted value of the API key121 The redacted value of the API key

122 122 

123 - `expires_at: optional number or null`

124 

125 The Unix timestamp (in seconds) when the API key expires, or null if it does not expire.

126 

123- `has_more: boolean`127- `has_more: boolean`

124 128 

125- `object: "list"`129- `object: "list"`

165 }169 }

166 },170 },

167 "owner_project_access": "active",171 "owner_project_access": "active",

168 "redacted_value": "redacted_value"172 "redacted_value": "redacted_value",

173 "expires_at": 0

169 }174 }

170 ],175 ],

171 "has_more": true,176 "has_more": true,

12 12 

13### Returns13### Returns

14 14 

15- `ProjectAPIKey object { id, created_at, last_used_at, 5 more }`15- `ProjectAPIKey object { id, created_at, last_used_at, 6 more }`

16 16 

17 Represents an individual API key in a project.17 Represents an individual API key in a project.

18 18 

104 104 

105 The redacted value of the API key105 The redacted value of the API key

106 106 

107 - `expires_at: optional number or null`

108 

109 The Unix timestamp (in seconds) when the API key expires, or null if it does not expire.

110 

107### Example111### Example

108 112 

109```http113```http

137 }141 }

138 },142 },

139 "owner_project_access": "active",143 "owner_project_access": "active",

140 "redacted_value": "redacted_value"144 "redacted_value": "redacted_value",

145 "expires_at": 0

141}146}

142```147```

143 148 

18 18 

19 Create the service account without default roles or an API key.19 Create the service account without default roles or an API key.

20 20 

21- `expires_in_seconds: optional number or null`

22 

23 Number of seconds until the initial API key expires. If omitted or null, the key does not expire unless the effective organization or project policy requires an expiration. When a policy sets a maximum lifetime, this value must be provided and must not exceed that limit. A non-null value cannot be used when `create_service_account_only` is true.

24 

21### Returns25### Returns

22 26 

23- `id: string`27- `id: string`

24 28 

25- `api_key: object { id, created_at, name, 2 more } or null`29- `api_key: object { id, created_at, name, 3 more } or null`

26 30 

27 - `id: string`31 - `id: string`

28 32 

38 42 

39 - `value: string`43 - `value: string`

40 44 

45 - `expires_at: optional number or null`

46 

47 The Unix timestamp (in seconds) when the API key expires, or null if it does not expire.

48 

41- `created_at: number`49- `created_at: number`

42 50 

43- `name: string`51- `name: string`

75 "created_at": 0,83 "created_at": 0,

76 "name": "name",84 "name": "name",

77 "object": "organization.project.service_account.api_key",85 "object": "organization.project.service_account.api_key",

78 "value": "value"86 "value": "value",

87 "expires_at": 0

79 },88 },

80 "created_at": 0,89 "created_at": 0,

81 "name": "name",90 "name": "name",

14 14 

15- `model: string or SpeechModel`15- `model: string or SpeechModel`

16 16 

17 One of the available [TTS models](/docs/models#tts): `tts-1`, `tts-1-hd`, `gpt-4o-mini-tts`, or `gpt-4o-mini-tts-2025-12-15`.17 One of the available [TTS models](/api/docs/guides/text-to-speech): `tts-1`, `tts-1-hd`, `gpt-4o-mini-tts`, or `gpt-4o-mini-tts-2025-12-15`.

18 18 

19 - `string`19 - `string`

20 20 

30 30 

31- `voice: string or "alloy" or "ash" or "ballad" or 7 more or object { id }`31- `voice: string or "alloy" or "ash" or "ballad" or 7 more or object { id }`

32 32 

33 The voice to use when generating the audio. Supported built-in voices are `alloy`, `ash`, `ballad`, `coral`, `echo`, `fable`, `onyx`, `nova`, `sage`, `shimmer`, `verse`, `marin`, and `cedar`. You may also provide a custom voice object with an `id`, for example `{ "id": "voice_1234" }`. Previews of the voices are available in the [Text to speech guide](/docs/guides/text-to-speech#voice-options).33 The voice to use when generating the audio. Supported built-in voices are `alloy`, `ash`, `ballad`, `coral`, `echo`, `fable`, `onyx`, `nova`, `sage`, `shimmer`, `verse`, `marin`, and `cedar`. You may also provide a custom voice object with an `id`, for example `{ "id": "voice_1234" }`. Previews of the voices are available in the [Text to speech guide](/api/docs/guides/text-to-speech#voice-options).

34 34 

35 - `string`35 - `string`

36 36 

4 4 

5## transcript.text.segment5## transcript.text.segment

6 6 

7Emitted when a diarized transcription returns a completed segment with speaker information. Only emitted when you [create a transcription](https://developers.openai.com/docs/api-reference/audio/create-transcription) with `stream` set to `true` and `response_format` set to `diarized_json`.7Emitted when a diarized transcription returns a completed segment with speaker information. Only emitted when you [create a transcription](https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create) with `stream` set to `true` and `response_format` set to `diarized_json`.

8 8 

9### Schema9### Schema

10 10 

15 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema)": {15 "(resource) audio.transcriptions > (model) transcription_text_segment_event > (schema)": {

16 "kind": "HttpDeclTypeAlias",16 "kind": "HttpDeclTypeAlias",

17 "oasRef": "#/components/schemas/TranscriptTextSegmentEvent",17 "oasRef": "#/components/schemas/TranscriptTextSegmentEvent",

18 "docstring": "Emitted when a diarized transcription returns a completed segment with speaker information. Only emitted when you [create a transcription](/docs/api-reference/audio/create-transcription) with `stream` set to `true` and `response_format` set to `diarized_json`.\n",18 "docstring": "Emitted when a diarized transcription returns a completed segment with speaker information. Only emitted when you [create a transcription](/api/reference/resources/audio/subresources/transcriptions/methods/create) with `stream` set to `true` and `response_format` set to `diarized_json`.\n",

19 "ident": "TranscriptionTextSegmentEvent",19 "ident": "TranscriptionTextSegmentEvent",

20 "type": {20 "type": {

21 "kind": "HttpTypeObject",21 "kind": "HttpTypeObject",

175 175 

176## transcript.text.delta176## transcript.text.delta

177 177 

178Emitted when there is an additional text delta. This is also the first event emitted when the transcription starts. Only emitted when you [create a transcription](https://developers.openai.com/docs/api-reference/audio/create-transcription) with the `Stream` parameter set to `true`.178Emitted when there is an additional text delta. This is also the first event emitted when the transcription starts. Only emitted when you [create a transcription](https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create) with the `Stream` parameter set to `true`.

179 179 

180### Schema180### Schema

181 181 

186 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema)": {186 "(resource) audio.transcriptions > (model) transcription_text_delta_event > (schema)": {

187 "kind": "HttpDeclTypeAlias",187 "kind": "HttpDeclTypeAlias",

188 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent",188 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent",

189 "docstring": "Emitted when there is an additional text delta. This is also the first event emitted when the transcription starts. Only emitted when you [create a transcription](/docs/api-reference/audio/create-transcription) with the `Stream` parameter set to `true`.",189 "docstring": "Emitted when there is an additional text delta. This is also the first event emitted when the transcription starts. Only emitted when you [create a transcription](/api/reference/resources/audio/subresources/transcriptions/methods/create) with the `Stream` parameter set to `true`.",

190 "ident": "TranscriptionTextDeltaEvent",190 "ident": "TranscriptionTextDeltaEvent",

191 "type": {191 "type": {

192 "kind": "HttpTypeObject",192 "kind": "HttpTypeObject",

256 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs",256 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs",

257 "deprecated": false,257 "deprecated": false,

258 "key": "logprobs",258 "key": "logprobs",

259 "docstring": "The log probabilities of the delta. Only included if you [create a transcription](/docs/api-reference/audio/create-transcription) with the `include[]` parameter set to `logprobs`.\n",259 "docstring": "The log probabilities of the delta. Only included if you [create a transcription](/api/reference/resources/audio/subresources/transcriptions/methods/create) with the `include[]` parameter set to `logprobs`.\n",

260 "type": {260 "type": {

261 "kind": "HttpTypeArray",261 "kind": "HttpTypeArray",

262 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs",262 "oasRef": "#/components/schemas/TranscriptTextDeltaEvent/properties/logprobs",

366 366 

367## transcript.text.done367## transcript.text.done

368 368 

369Emitted when the transcription is complete. Contains the complete transcription text. Only emitted when you [create a transcription](https://developers.openai.com/docs/api-reference/audio/create-transcription) with the `Stream` parameter set to `true`.369Emitted when the transcription is complete. Contains the complete transcription text. Only emitted when you [create a transcription](https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create) with the `Stream` parameter set to `true`.

370 370 

371### Schema371### Schema

372 372 

377 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema)": {377 "(resource) audio.transcriptions > (model) transcription_text_done_event > (schema)": {

378 "kind": "HttpDeclTypeAlias",378 "kind": "HttpDeclTypeAlias",

379 "oasRef": "#/components/schemas/TranscriptTextDoneEvent",379 "oasRef": "#/components/schemas/TranscriptTextDoneEvent",

380 "docstring": "Emitted when the transcription is complete. Contains the complete transcription text. Only emitted when you [create a transcription](/docs/api-reference/audio/create-transcription) with the `Stream` parameter set to `true`.",380 "docstring": "Emitted when the transcription is complete. Contains the complete transcription text. Only emitted when you [create a transcription](/api/reference/resources/audio/subresources/transcriptions/methods/create) with the `Stream` parameter set to `true`.",

381 "ident": "TranscriptionTextDoneEvent",381 "ident": "TranscriptionTextDoneEvent",

382 "type": {382 "type": {

383 "kind": "HttpTypeObject",383 "kind": "HttpTypeObject",

474 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs",474 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs",

475 "deprecated": false,475 "deprecated": false,

476 "key": "logprobs",476 "key": "logprobs",

477 "docstring": "The log probabilities of the individual tokens in the transcription. Only included if you [create a transcription](/docs/api-reference/audio/create-transcription) with the `include[]` parameter set to `logprobs`.\n",477 "docstring": "The log probabilities of the individual tokens in the transcription. Only included if you [create a transcription](/api/reference/resources/audio/subresources/transcriptions/methods/create) with the `include[]` parameter set to `logprobs`.\n",

478 "type": {478 "type": {

479 "kind": "HttpTypeArray",479 "kind": "HttpTypeArray",

480 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs",480 "oasRef": "#/components/schemas/TranscriptTextDoneEvent/properties/logprobs",

130 Model ID used to process the batch, like `gpt-6-astra`. OpenAI130 Model ID used to process the batch, like `gpt-6-astra`. OpenAI

131 offers a wide range of models with different capabilities, performance131 offers a wide range of models with different capabilities, performance

132 characteristics, and price points. Refer to the [model132 characteristics, and price points. Refer to the [model

133 guide](/docs/models) to browse and compare available models.133 guide](/api/docs/models) to browse and compare available models.

134 134 

135 - `output_file_id: optional string`135 - `output_file_id: optional string`

136 136 

169 - `cached_tokens: number`169 - `cached_tokens: number`

170 170 

171 The number of tokens that were retrieved from the cache. [More on171 The number of tokens that were retrieved from the cache. [More on

172 prompt caching](/docs/guides/prompt-caching).172 prompt caching](/api/docs/guides/prompt-caching).

173 173 

174 - `output_tokens: number`174 - `output_tokens: number`

175 175 

36 36 

37 The ID of an uploaded file that contains requests for the new batch.37 The ID of an uploaded file that contains requests for the new batch.

38 38 

39 See [upload file](/docs/api-reference/files/create) for how to upload a file.39 See [upload file](/api/reference/resources/files/methods/create) for how to upload a file.

40 40 

41 Your input file must be formatted as a [JSONL file](/docs/api-reference/batch/request-input), and must be uploaded with the purpose `batch`. The file can contain up to 50,000 requests, and can be up to 200 MB in size.41 Your input file must be formatted as a [JSONL file](/api/docs/guides/batch#1-prepare-your-batch-file), and must be uploaded with the purpose `batch`. The file can contain up to 50,000 requests, and can be up to 200 MB in size.

42 42 

43- `metadata: optional Metadata or null`43- `metadata: optional Metadata or null`

44 44 

185 Model ID used to process the batch, like `gpt-6-astra`. OpenAI185 Model ID used to process the batch, like `gpt-6-astra`. OpenAI

186 offers a wide range of models with different capabilities, performance186 offers a wide range of models with different capabilities, performance

187 characteristics, and price points. Refer to the [model187 characteristics, and price points. Refer to the [model

188 guide](/docs/models) to browse and compare available models.188 guide](/api/docs/models) to browse and compare available models.

189 189 

190 - `output_file_id: optional string`190 - `output_file_id: optional string`

191 191 

224 - `cached_tokens: number`224 - `cached_tokens: number`

225 225 

226 The number of tokens that were retrieved from the cache. [More on226 The number of tokens that were retrieved from the cache. [More on

227 prompt caching](/docs/guides/prompt-caching).227 prompt caching](/api/docs/guides/prompt-caching).

228 228 

229 - `output_tokens: number`229 - `output_tokens: number`

230 230 

136 Model ID used to process the batch, like `gpt-6-astra`. OpenAI136 Model ID used to process the batch, like `gpt-6-astra`. OpenAI

137 offers a wide range of models with different capabilities, performance137 offers a wide range of models with different capabilities, performance

138 characteristics, and price points. Refer to the [model138 characteristics, and price points. Refer to the [model

139 guide](/docs/models) to browse and compare available models.139 guide](/api/docs/models) to browse and compare available models.

140 140 

141 - `output_file_id: optional string`141 - `output_file_id: optional string`

142 142 

175 - `cached_tokens: number`175 - `cached_tokens: number`

176 176 

177 The number of tokens that were retrieved from the cache. [More on177 The number of tokens that were retrieved from the cache. [More on

178 prompt caching](/docs/guides/prompt-caching).178 prompt caching](/api/docs/guides/prompt-caching).

179 179 

180 - `output_tokens: number`180 - `output_tokens: number`

181 181 

130 Model ID used to process the batch, like `gpt-6-astra`. OpenAI130 Model ID used to process the batch, like `gpt-6-astra`. OpenAI

131 offers a wide range of models with different capabilities, performance131 offers a wide range of models with different capabilities, performance

132 characteristics, and price points. Refer to the [model132 characteristics, and price points. Refer to the [model

133 guide](/docs/models) to browse and compare available models.133 guide](/api/docs/models) to browse and compare available models.

134 134 

135 - `output_file_id: optional string`135 - `output_file_id: optional string`

136 136 

169 - `cached_tokens: number`169 - `cached_tokens: number`

170 170 

171 The number of tokens that were retrieved from the cache. [More on171 The number of tokens that were retrieved from the cache. [More on

172 prompt caching](/docs/guides/prompt-caching).172 prompt caching](/api/docs/guides/prompt-caching).

173 173 

174 - `output_tokens: number`174 - `output_tokens: number`

175 175 

4 4 

5Cancels a model response with the given ID. Only responses created with5Cancels a model response with the given ID. Only responses created with

6the `background` parameter set to `true` can be cancelled.6the `background` parameter set to `true` can be cancelled.

7[Learn more](/docs/guides/background).7[Learn more](/api/docs/guides/background).

8 8 

9### Header Parameters9### Header Parameters

10 10 

199 199 

200 - `BetaResponseInputImage object { detail, type, file_id, 2 more }`200 - `BetaResponseInputImage object { detail, type, file_id, 2 more }`

201 201 

202 An image input to the model. Learn about [image inputs](/docs/guides/vision).202 An image input to the model. Learn about [image inputs](/api/docs/guides/images-vision).

203 203 

204 - `detail: BetaImageDetail`204 - `detail: BetaImageDetail`

205 205 

558 - `FileSearchCall object { id, queries, status, 3 more }`558 - `FileSearchCall object { id, queries, status, 3 more }`

559 559 

560 The results of a file search tool call. See the560 The results of a file search tool call. See the

561 [file search guide](/docs/guides/tools-file-search) for more information.561 [file search guide](/api/docs/guides/tools-file-search) for more information.

562 562 

563 - `id: string`563 - `id: string`

564 564 

634 - `ComputerCall object { id, call_id, pending_safety_checks, 5 more }`634 - `ComputerCall object { id, call_id, pending_safety_checks, 5 more }`

635 635 

636 A tool call to a computer use tool. See the636 A tool call to a computer use tool. See the

637 [computer use guide](/docs/guides/tools-computer-use) for more information.637 [computer use guide](/api/docs/guides/tools-computer-use) for more information.

638 638 

639 - `id: string`639 - `id: string`

640 640 

995 - `WebSearchCall object { id, action, status, 2 more }`995 - `WebSearchCall object { id, action, status, 2 more }`

996 996 

997 The results of a web search tool call. See the997 The results of a web search tool call. See the

998 [web search guide](/docs/guides/tools-web-search) for more information.998 [web search guide](/api/docs/guides/tools-web-search) for more information.

999 999 

1000 - `id: string`1000 - `id: string`

1001 1001 

1101 - `FunctionCall object { arguments, call_id, name, 7 more }`1101 - `FunctionCall object { arguments, call_id, name, 7 more }`

1102 1102 

1103 A tool call to run a function. See the1103 A tool call to run a function. See the

1104 [function calling guide](/docs/guides/function-calling) for more information.1104 [function calling guide](/api/docs/guides/function-calling) for more information.

1105 1105 

1106 - `arguments: string`1106 - `arguments: string`

1107 1107 

1214 1214 

1215 - `BetaResponseInputImageContent object { type, detail, file_id, 2 more }`1215 - `BetaResponseInputImageContent object { type, detail, file_id, 2 more }`

1216 1216 

1217 An image input to the model. Learn about [image inputs](/docs/guides/vision)1217 An image input to the model. Learn about [image inputs](/api/docs/guides/images-vision)

1218 1218 

1219 - `type: "input_image"`1219 - `type: "input_image"`

1220 1220 

1372 1372 

1373 - `BetaResponseInputImageContent object { type, detail, file_id, 2 more }`1373 - `BetaResponseInputImageContent object { type, detail, file_id, 2 more }`

1374 1374 

1375 An image input to the model. Learn about [image inputs](/docs/guides/vision)1375 An image input to the model. Learn about [image inputs](/api/docs/guides/images-vision)

1376 1376 

1377 - `EncryptedContent object { encrypted_content, type }`1377 - `EncryptedContent object { encrypted_content, type }`

1378 1378 

1638 1638 

1639 - `Function object { name, parameters, strict, 6 more }`1639 - `Function object { name, parameters, strict, 6 more }`

1640 1640 

1641 Defines a function in your own code the model can choose to call. Learn more about [function calling](https://platform.openai.com/docs/guides/function-calling).1641 Defines a function in your own code the model can choose to call. Learn more about [function calling](/api/docs/guides/function-calling).

1642 1642 

1643 - `name: string`1643 - `name: string`

1644 1644 

1682 1682 

1683 - `FileSearch object { type, vector_store_ids, filters, 2 more }`1683 - `FileSearch object { type, vector_store_ids, filters, 2 more }`

1684 1684 

1685 A tool that searches for relevant content from uploaded files. Learn more about the [file search tool](https://platform.openai.com/docs/guides/tools-file-search).1685 A tool that searches for relevant content from uploaded files. Learn more about the [file search tool](/api/docs/guides/tools-file-search).

1686 1686 

1687 - `type: "file_search"`1687 - `type: "file_search"`

1688 1688 

1856 1856 

1857 - `Computer object { type }`1857 - `Computer object { type }`

1858 1858 

1859 A tool that controls a virtual computer. Learn more about the [computer tool](https://platform.openai.com/docs/guides/tools-computer-use).1859 A tool that controls a virtual computer. Learn more about the [computer tool](/api/docs/guides/tools-computer-use).

1860 1860 

1861 - `type: "computer"`1861 - `type: "computer"`

1862 1862 

1866 1866 

1867 - `ComputerUsePreview object { display_height, display_width, environment, type }`1867 - `ComputerUsePreview object { display_height, display_width, environment, type }`

1868 1868 

1869 A tool that controls a virtual computer. Learn more about the [computer tool](https://platform.openai.com/docs/guides/tools-computer-use).1869 A tool that controls a virtual computer. Learn more about the [computer tool](/api/docs/guides/tools-computer-use).

1870 1870 

1871 - `display_height: number`1871 - `display_height: number`

1872 1872 

1899 - `WebSearch object { type, external_web_access, filters, 2 more }`1899 - `WebSearch object { type, external_web_access, filters, 2 more }`

1900 1900 

1901 Search the Internet for sources related to the prompt. Learn more about the1901 Search the Internet for sources related to the prompt. Learn more about the

1902 [web search tool](/docs/guides/tools-web-search).1902 [web search tool](/api/docs/guides/tools-web-search).

1903 1903 

1904 - `type: "web_search" or "web_search_2025_08_26"`1904 - `type: "web_search" or "web_search_2025_08_26"`

1905 1905 

1963 - `Mcp object { server_label, type, allowed_callers, 9 more }`1963 - `Mcp object { server_label, type, allowed_callers, 9 more }`

1964 1964 

1965 Give the model access to additional tools via remote Model Context Protocol1965 Give the model access to additional tools via remote Model Context Protocol

1966 (MCP) servers. [Learn more about MCP](/docs/guides/tools-remote-mcp).1966 (MCP) servers. [Learn more about MCP](/api/docs/guides/tools-connectors-mcp).

1967 1967 

1968 - `server_label: string`1968 - `server_label: string`

1969 1969 

2015 2015 

2016 Identifier for service connectors, like those available in ChatGPT. One of2016 Identifier for service connectors, like those available in ChatGPT. One of

2017 `server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more2017 `server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more

2018 about service connectors [here](/docs/guides/tools-remote-mcp#connectors).2018 about service connectors [here](/api/docs/guides/tools-connectors-mcp#connectors).

2019 2019 

2020 Currently supported `connector_id` values are:2020 Currently supported `connector_id` values are:

2021 2021 

2239 2239 

2240 - `background: optional "transparent" or "opaque" or "auto"`2240 - `background: optional "transparent" or "opaque" or "auto"`

2241 2241 

2242 Set the background of the generated image. One of `transparent`,2242 Set the background of the generated image. One of `transparent`, `opaque`,

2243 `opaque`, or `auto`. Transparent backgrounds are available for2243 or `auto`. `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare`, including

2244 supported GPT Image models. For `gpt-image-2` and2244 their `2026-09-08` snapshots, support `opaque` and `transparent`

2245 `gpt-image-2-2026-04-21`, this support is in preview. When using2245 backgrounds. Transparent backgrounds are available for supported GPT Image

2246 `transparent`, set the output format to `png` or `webp`. Default: `auto`.2246 models. For `gpt-image-2` and `gpt-image-2-2026-04-21`, this support is in

2247 preview. When using `transparent`, set the output format to `png` or `webp`.

2248 Default: `auto`.

2247 2249 

2248 - `"transparent"`2250 - `"transparent"`

2249 2251 

2272 2274 

2273 Base64-encoded mask image.2275 Base64-encoded mask image.

2274 2276 

2275 - `model: optional string or "gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 2 more`2277 - `model: optional string or "gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 6 more`

2276 2278 

2277 The image generation model to use. One of `gpt-image-1`,2279 The image generation model to use. One of `gpt-image-1`,

2278 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,2280 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,

2279 `gpt-image-2-2026-04-21`, or `chatgpt-image-latest`. Default:2281 `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`,

2282 `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`,

2283 `gpt-image-2.5-flare-2026-09-08`, or `chatgpt-image-latest`. Default:

2280 `gpt-image-1`.2284 `gpt-image-1`.

2281 2285 

2282 - `string`2286 - `string`

2283 2287 

2284 - `"gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 2 more`2288 - `"gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 6 more`

2285 2289 

2286 The image generation model to use. One of `gpt-image-1`,2290 The image generation model to use. One of `gpt-image-1`,

2287 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,2291 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,

2288 `gpt-image-2-2026-04-21`, or `chatgpt-image-latest`. Default:2292 `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`,

2293 `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`,

2294 `gpt-image-2.5-flare-2026-09-08`, or `chatgpt-image-latest`. Default:

2289 `gpt-image-1`.2295 `gpt-image-1`.

2290 2296 

2291 - `"gpt-image-1"`2297 - `"gpt-image-1"`

2298 2304 

2299 - `"gpt-image-2-2026-04-21"`2305 - `"gpt-image-2-2026-04-21"`

2300 2306 

2307 - `"gpt-image-2.5-sunburst"`

2308 

2309 - `"gpt-image-2.5-sunburst-2026-09-08"`

2310 

2311 - `"gpt-image-2.5-flare"`

2312 

2313 - `"gpt-image-2.5-flare-2026-09-08"`

2314 

2301 - `moderation: optional "auto" or "low"`2315 - `moderation: optional "auto" or "low"`

2302 2316 

2303 Moderation level for the generated image. Default: `auto`.2317 Moderation level for the generated image. Default: `auto`.

2325 2339 

2326 Number of partial images to generate in streaming mode, from 0 (default value) to 3.2340 Number of partial images to generate in streaming mode, from 0 (default value) to 3.

2327 2341 

2328 - `quality: optional "low" or "medium" or "high" or "auto"`2342 - `quality: optional "low" or "medium" or "high" or 3 more`

2329 2343 

2330 The quality of the generated image. One of `low`, `medium`, `high`,2344 The quality of the generated image. The GPT image models support `low`,

2331 or `auto`. Default: `auto`.2345 `medium`, and `high`. `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare`,

2346 including their `2026-09-08` snapshots, also support `xhigh` and `max`.

2347 Default: `auto`.

2332 2348 

2333 - `"low"`2349 - `"low"`

2334 2350 

2336 2352 

2337 - `"high"`2353 - `"high"`

2338 2354 

2355 - `"xhigh"`

2356 

2357 - `"max"`

2358 

2339 - `"auto"`2359 - `"auto"`

2340 2360 

2341 - `size: optional string or "1024x1024" or "1024x1536" or "1536x1024" or "auto"`2361 - `size: optional string or "1024x1024" or "1024x1536" or "1536x1024" or "auto"`

2342 2362 

2343 The size of the generated images. For `gpt-image-2` and `gpt-image-2-2026-04-21`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.2363 The size of the generated images. For `gpt-image-2`, `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`, `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`, and `gpt-image-2.5-flare-2026-09-08`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.

2344 2364 

2345 - `string`2365 - `string`

2346 2366 

2347 - `"1024x1024" or "1024x1536" or "1536x1024" or "auto"`2367 - `"1024x1024" or "1024x1536" or "1536x1024" or "auto"`

2348 2368 

2349 The size of the generated images. For `gpt-image-2` and `gpt-image-2-2026-04-21`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.2369 The size of the generated images. For `gpt-image-2`, `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`, `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`, and `gpt-image-2.5-flare-2026-09-08`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.

2350 2370 

2351 - `"1024x1024"`2371 - `"1024x1024"`

2352 2372 

2512 2532 

2513 - `Custom object { name, type, allowed_callers, 4 more }`2533 - `Custom object { name, type, allowed_callers, 4 more }`

2514 2534 

2515 A custom tool that processes input using a specified format. Learn more about [custom tools](/docs/guides/function-calling#custom-tools)2535 A custom tool that processes input using a specified format. Learn more about [custom tools](/api/docs/guides/function-calling#custom-tools)

2516 2536 

2517 - `name: string`2537 - `name: string`

2518 2538 

2634 2654 

2635 - `Custom object { name, type, allowed_callers, 4 more }`2655 - `Custom object { name, type, allowed_callers, 4 more }`

2636 2656 

2637 A custom tool that processes input using a specified format. Learn more about [custom tools](/docs/guides/function-calling#custom-tools)2657 A custom tool that processes input using a specified format. Learn more about [custom tools](/api/docs/guides/function-calling#custom-tools)

2638 2658 

2639 - `name: string`2659 - `name: string`

2640 2660 

2736 2756 

2737 - `WebSearchPreview object { type, search_content_types, search_context_size, user_location }`2757 - `WebSearchPreview object { type, search_content_types, search_context_size, user_location }`

2738 2758 

2739 This tool searches the web for relevant results to use in a response. Learn more about the [web search tool](https://platform.openai.com/docs/guides/tools-web-search).2759 This tool searches the web for relevant results to use in a response. Learn more about the [web search tool](/api/docs/guides/tools-web-search).

2740 2760 

2741 - `type: "web_search_preview" or "web_search_preview_2025_03_11"`2761 - `type: "web_search_preview" or "web_search_preview_2025_03_11"`

2742 2762 

2860 2880 

2861 - `Function object { name, parameters, strict, 6 more }`2881 - `Function object { name, parameters, strict, 6 more }`

2862 2882 

2863 Defines a function in your own code the model can choose to call. Learn more about [function calling](https://platform.openai.com/docs/guides/function-calling).2883 Defines a function in your own code the model can choose to call. Learn more about [function calling](/api/docs/guides/function-calling).

2864 2884 

2865 - `name: string`2885 - `name: string`

2866 2886 

2904 2924 

2905 - `FileSearch object { type, vector_store_ids, filters, 2 more }`2925 - `FileSearch object { type, vector_store_ids, filters, 2 more }`

2906 2926 

2907 A tool that searches for relevant content from uploaded files. Learn more about the [file search tool](https://platform.openai.com/docs/guides/tools-file-search).2927 A tool that searches for relevant content from uploaded files. Learn more about the [file search tool](/api/docs/guides/tools-file-search).

2908 2928 

2909 - `type: "file_search"`2929 - `type: "file_search"`

2910 2930 

3078 3098 

3079 - `Computer object { type }`3099 - `Computer object { type }`

3080 3100 

3081 A tool that controls a virtual computer. Learn more about the [computer tool](https://platform.openai.com/docs/guides/tools-computer-use).3101 A tool that controls a virtual computer. Learn more about the [computer tool](/api/docs/guides/tools-computer-use).

3082 3102 

3083 - `type: "computer"`3103 - `type: "computer"`

3084 3104 

3088 3108 

3089 - `ComputerUsePreview object { display_height, display_width, environment, type }`3109 - `ComputerUsePreview object { display_height, display_width, environment, type }`

3090 3110 

3091 A tool that controls a virtual computer. Learn more about the [computer tool](https://platform.openai.com/docs/guides/tools-computer-use).3111 A tool that controls a virtual computer. Learn more about the [computer tool](/api/docs/guides/tools-computer-use).

3092 3112 

3093 - `display_height: number`3113 - `display_height: number`

3094 3114 

3121 - `WebSearch object { type, external_web_access, filters, 2 more }`3141 - `WebSearch object { type, external_web_access, filters, 2 more }`

3122 3142 

3123 Search the Internet for sources related to the prompt. Learn more about the3143 Search the Internet for sources related to the prompt. Learn more about the

3124 [web search tool](/docs/guides/tools-web-search).3144 [web search tool](/api/docs/guides/tools-web-search).

3125 3145 

3126 - `type: "web_search" or "web_search_2025_08_26"`3146 - `type: "web_search" or "web_search_2025_08_26"`

3127 3147 

3185 - `Mcp object { server_label, type, allowed_callers, 9 more }`3205 - `Mcp object { server_label, type, allowed_callers, 9 more }`

3186 3206 

3187 Give the model access to additional tools via remote Model Context Protocol3207 Give the model access to additional tools via remote Model Context Protocol

3188 (MCP) servers. [Learn more about MCP](/docs/guides/tools-remote-mcp).3208 (MCP) servers. [Learn more about MCP](/api/docs/guides/tools-connectors-mcp).

3189 3209 

3190 - `server_label: string`3210 - `server_label: string`

3191 3211 

3237 3257 

3238 Identifier for service connectors, like those available in ChatGPT. One of3258 Identifier for service connectors, like those available in ChatGPT. One of

3239 `server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more3259 `server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more

3240 about service connectors [here](/docs/guides/tools-remote-mcp#connectors).3260 about service connectors [here](/api/docs/guides/tools-connectors-mcp#connectors).

3241 3261 

3242 Currently supported `connector_id` values are:3262 Currently supported `connector_id` values are:

3243 3263 

3429 3449 

3430 - `background: optional "transparent" or "opaque" or "auto"`3450 - `background: optional "transparent" or "opaque" or "auto"`

3431 3451 

3432 Set the background of the generated image. One of `transparent`,3452 Set the background of the generated image. One of `transparent`, `opaque`,

3433 `opaque`, or `auto`. Transparent backgrounds are available for3453 or `auto`. `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare`, including

3434 supported GPT Image models. For `gpt-image-2` and3454 their `2026-09-08` snapshots, support `opaque` and `transparent`

3435 `gpt-image-2-2026-04-21`, this support is in preview. When using3455 backgrounds. Transparent backgrounds are available for supported GPT Image

3436 `transparent`, set the output format to `png` or `webp`. Default: `auto`.3456 models. For `gpt-image-2` and `gpt-image-2-2026-04-21`, this support is in

3457 preview. When using `transparent`, set the output format to `png` or `webp`.

3458 Default: `auto`.

3437 3459 

3438 - `"transparent"`3460 - `"transparent"`

3439 3461 

3462 3484 

3463 Base64-encoded mask image.3485 Base64-encoded mask image.

3464 3486 

3465 - `model: optional string or "gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 2 more`3487 - `model: optional string or "gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 6 more`

3466 3488 

3467 The image generation model to use. One of `gpt-image-1`,3489 The image generation model to use. One of `gpt-image-1`,

3468 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,3490 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,

3469 `gpt-image-2-2026-04-21`, or `chatgpt-image-latest`. Default:3491 `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`,

3492 `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`,

3493 `gpt-image-2.5-flare-2026-09-08`, or `chatgpt-image-latest`. Default:

3470 `gpt-image-1`.3494 `gpt-image-1`.

3471 3495 

3472 - `string`3496 - `string`

3473 3497 

3474 - `"gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 2 more`3498 - `"gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 6 more`

3475 3499 

3476 The image generation model to use. One of `gpt-image-1`,3500 The image generation model to use. One of `gpt-image-1`,

3477 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,3501 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,

3478 `gpt-image-2-2026-04-21`, or `chatgpt-image-latest`. Default:3502 `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`,

3503 `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`,

3504 `gpt-image-2.5-flare-2026-09-08`, or `chatgpt-image-latest`. Default:

3479 `gpt-image-1`.3505 `gpt-image-1`.

3480 3506 

3481 - `"gpt-image-1"`3507 - `"gpt-image-1"`

3488 3514 

3489 - `"gpt-image-2-2026-04-21"`3515 - `"gpt-image-2-2026-04-21"`

3490 3516 

3517 - `"gpt-image-2.5-sunburst"`

3518 

3519 - `"gpt-image-2.5-sunburst-2026-09-08"`

3520 

3521 - `"gpt-image-2.5-flare"`

3522 

3523 - `"gpt-image-2.5-flare-2026-09-08"`

3524 

3491 - `moderation: optional "auto" or "low"`3525 - `moderation: optional "auto" or "low"`

3492 3526 

3493 Moderation level for the generated image. Default: `auto`.3527 Moderation level for the generated image. Default: `auto`.

3515 3549 

3516 Number of partial images to generate in streaming mode, from 0 (default value) to 3.3550 Number of partial images to generate in streaming mode, from 0 (default value) to 3.

3517 3551 

3518 - `quality: optional "low" or "medium" or "high" or "auto"`3552 - `quality: optional "low" or "medium" or "high" or 3 more`

3519 3553 

3520 The quality of the generated image. One of `low`, `medium`, `high`,3554 The quality of the generated image. The GPT image models support `low`,

3521 or `auto`. Default: `auto`.3555 `medium`, and `high`. `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare`,

3556 including their `2026-09-08` snapshots, also support `xhigh` and `max`.

3557 Default: `auto`.

3522 3558 

3523 - `"low"`3559 - `"low"`

3524 3560 

3526 3562 

3527 - `"high"`3563 - `"high"`

3528 3564 

3565 - `"xhigh"`

3566 

3567 - `"max"`

3568 

3529 - `"auto"`3569 - `"auto"`

3530 3570 

3531 - `size: optional string or "1024x1024" or "1024x1536" or "1536x1024" or "auto"`3571 - `size: optional string or "1024x1024" or "1024x1536" or "1536x1024" or "auto"`

3532 3572 

3533 The size of the generated images. For `gpt-image-2` and `gpt-image-2-2026-04-21`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.3573 The size of the generated images. For `gpt-image-2`, `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`, `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`, and `gpt-image-2.5-flare-2026-09-08`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.

3534 3574 

3535 - `string`3575 - `string`

3536 3576 

3537 - `"1024x1024" or "1024x1536" or "1536x1024" or "auto"`3577 - `"1024x1024" or "1024x1536" or "1536x1024" or "auto"`

3538 3578 

3539 The size of the generated images. For `gpt-image-2` and `gpt-image-2-2026-04-21`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.3579 The size of the generated images. For `gpt-image-2`, `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`, `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`, and `gpt-image-2.5-flare-2026-09-08`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.

3540 3580 

3541 - `"1024x1024"`3581 - `"1024x1024"`

3542 3582 

3584 3624 

3585 - `Custom object { name, type, allowed_callers, 4 more }`3625 - `Custom object { name, type, allowed_callers, 4 more }`

3586 3626 

3587 A custom tool that processes input using a specified format. Learn more about [custom tools](/docs/guides/function-calling#custom-tools)3627 A custom tool that processes input using a specified format. Learn more about [custom tools](/api/docs/guides/function-calling#custom-tools)

3588 3628 

3589 - `name: string`3629 - `name: string`

3590 3630 

3706 3746 

3707 - `Custom object { name, type, allowed_callers, 4 more }`3747 - `Custom object { name, type, allowed_callers, 4 more }`

3708 3748 

3709 A custom tool that processes input using a specified format. Learn more about [custom tools](/docs/guides/function-calling#custom-tools)3749 A custom tool that processes input using a specified format. Learn more about [custom tools](/api/docs/guides/function-calling#custom-tools)

3710 3750 

3711 - `name: string`3751 - `name: string`

3712 3752 

3808 3848 

3809 - `WebSearchPreview object { type, search_content_types, search_context_size, user_location }`3849 - `WebSearchPreview object { type, search_content_types, search_context_size, user_location }`

3810 3850 

3811 This tool searches the web for relevant results to use in a response. Learn more about the [web search tool](https://platform.openai.com/docs/guides/tools-web-search).3851 This tool searches the web for relevant results to use in a response. Learn more about the [web search tool](/api/docs/guides/tools-web-search).

3812 3852 

3813 - `type: "web_search_preview" or "web_search_preview_2025_03_11"`3853 - `type: "web_search_preview" or "web_search_preview_2025_03_11"`

3814 3854 

3948 A description of the chain of thought used by a reasoning model while generating3988 A description of the chain of thought used by a reasoning model while generating

3949 a response. Be sure to include these items in your `input` to the Responses API3989 a response. Be sure to include these items in your `input` to the Responses API

3950 for subsequent turns of a conversation if you are manually3990 for subsequent turns of a conversation if you are manually

3951 [managing context](/docs/guides/conversation-state).3991 [managing context](/api/docs/guides/conversation-state).

3952 3992 

3953 - `id: string`3993 - `id: string`

3954 3994 

4021 4061 

4022 - `Compaction object { encrypted_content, type, id, agent }`4062 - `Compaction object { encrypted_content, type, id, agent }`

4023 4063 

4024 A compaction item generated by the [`v1/responses/compact` API](/docs/api-reference/responses/compact).4064 A compaction item generated by the [`v1/responses/compact` API](/api/reference/resources/responses/methods/compact).

4025 4065 

4026 - `encrypted_content: string`4066 - `encrypted_content: string`

4027 4067 

4045 4085 

4046 The canonical name of the agent that produced this item.4086 The canonical name of the agent that produced this item.

4047 4087 

4048 - `ImageGenerationCall object { id, result, status, 2 more }`4088 - `ImageGenerationCall object { id, result, status, 4 more }`

4049 4089 

4050 An image generation request made by the model.4090 An image generation request made by the model.

4051 4091 

4083 4123 

4084 The canonical name of the agent that produced this item.4124 The canonical name of the agent that produced this item.

4085 4125 

4126 - `quality: optional "low" or "medium" or "high" or 3 more or null`

4127 

4128 The quality of the image generated by the image generation tool call. One of `low`, `medium`, `high`, `xhigh`, `max`, or `auto`.

4129 

4130 - `"low"`

4131 

4132 - `"medium"`

4133 

4134 - `"high"`

4135 

4136 - `"xhigh"`

4137 

4138 - `"max"`

4139 

4140 - `"auto"`

4141 

4142 - `size: optional string or "1024x1024" or "1024x1536" or "1536x1024" or null`

4143 

4144 The image dimensions as a `WIDTHxHEIGHT` string, for example `1536x864`.

4145 

4146 - `string`

4147 

4148 - `"1024x1024" or "1024x1536" or "1536x1024"`

4149 

4150 The image dimensions as a `WIDTHxHEIGHT` string, for example `1536x864`.

4151 

4152 - `"1024x1024"`

4153 

4154 - `"1024x1536"`

4155 

4156 - `"1536x1024"`

4157 

4086 - `CodeInterpreterCall object { id, code, container_id, 4 more }`4158 - `CodeInterpreterCall object { id, code, container_id, 4 more }`

4087 4159 

4088 A tool call to run code.4160 A tool call to run code.

4858 4930 

4859 - `BetaResponseInputImage object { detail, type, file_id, 2 more }`4931 - `BetaResponseInputImage object { detail, type, file_id, 2 more }`

4860 4932 

4861 An image input to the model. Learn about [image inputs](/docs/guides/vision).4933 An image input to the model. Learn about [image inputs](/api/docs/guides/images-vision).

4862 4934 

4863 - `BetaResponseInputFile object { type, detail, file_data, 4 more }`4935 - `BetaResponseInputFile object { type, detail, file_data, 4 more }`

4864 4936 

5093 5165 

5094 Model ID used to generate the response, like `gpt-6-astra`. OpenAI5166 Model ID used to generate the response, like `gpt-6-astra`. OpenAI

5095 offers a wide range of models with different capabilities, performance5167 offers a wide range of models with different capabilities, performance

5096 characteristics, and price points. Refer to the [model guide](/docs/models)5168 characteristics, and price points. Refer to the [model guide](/api/docs/models)

5097 to browse and compare available models.5169 to browse and compare available models.

5098 5170 

5099 - `"gpt-6-astra" or "gpt-5.6-sol" or "gpt-5.6-terra" or 100 more`5171 - `"gpt-6-astra" or "gpt-5.6-sol" or "gpt-5.6-terra" or 100 more`

5100 5172 

5101 Model ID used to generate the response, like `gpt-6-astra`. OpenAI5173 Model ID used to generate the response, like `gpt-6-astra`. OpenAI

5102 offers a wide range of models with different capabilities, performance5174 offers a wide range of models with different capabilities, performance

5103 characteristics, and price points. Refer to the [model guide](/docs/models)5175 characteristics, and price points. Refer to the [model guide](/api/docs/models)

5104 to browse and compare available models.5176 to browse and compare available models.

5105 5177 

5106 - `"gpt-6-astra"`5178 - `"gpt-6-astra"`

5335 - `FileSearchCall object { id, queries, status, 3 more }`5407 - `FileSearchCall object { id, queries, status, 3 more }`

5336 5408 

5337 The results of a file search tool call. See the5409 The results of a file search tool call. See the

5338 [file search guide](/docs/guides/tools-file-search) for more information.5410 [file search guide](/api/docs/guides/tools-file-search) for more information.

5339 5411 

5340 - `id: string`5412 - `id: string`

5341 5413 

5411 - `FunctionCall object { arguments, call_id, name, 7 more }`5483 - `FunctionCall object { arguments, call_id, name, 7 more }`

5412 5484 

5413 A tool call to run a function. See the5485 A tool call to run a function. See the

5414 [function calling guide](/docs/guides/function-calling) for more information.5486 [function calling guide](/api/docs/guides/function-calling) for more information.

5415 5487 

5416 - `arguments: string`5488 - `arguments: string`

5417 5489 

5507 5579 

5508 - `BetaResponseInputImage object { detail, type, file_id, 2 more }`5580 - `BetaResponseInputImage object { detail, type, file_id, 2 more }`

5509 5581 

5510 An image input to the model. Learn about [image inputs](/docs/guides/vision).5582 An image input to the model. Learn about [image inputs](/api/docs/guides/images-vision).

5511 5583 

5512 - `BetaResponseInputFile object { type, detail, file_data, 4 more }`5584 - `BetaResponseInputFile object { type, detail, file_data, 4 more }`

5513 5585 

5644 5716 

5645 - `BetaResponseInputImage object { detail, type, file_id, 2 more }`5717 - `BetaResponseInputImage object { detail, type, file_id, 2 more }`

5646 5718 

5647 An image input to the model. Learn about [image inputs](/docs/guides/vision).5719 An image input to the model. Learn about [image inputs](/api/docs/guides/images-vision).

5648 5720 

5649 - `ComputerScreenshot object { detail, file_id, image_url, 2 more }`5721 - `ComputerScreenshot object { detail, file_id, image_url, 2 more }`

5650 5722 

5819 - `WebSearchCall object { id, action, status, 2 more }`5891 - `WebSearchCall object { id, action, status, 2 more }`

5820 5892 

5821 The results of a web search tool call. See the5893 The results of a web search tool call. See the

5822 [web search guide](/docs/guides/tools-web-search) for more information.5894 [web search guide](/api/docs/guides/tools-web-search) for more information.

5823 5895 

5824 - `id: string`5896 - `id: string`

5825 5897 

5925 - `ComputerCall object { id, call_id, pending_safety_checks, 5 more }`5997 - `ComputerCall object { id, call_id, pending_safety_checks, 5 more }`

5926 5998 

5927 A tool call to a computer use tool. See the5999 A tool call to a computer use tool. See the

5928 [computer use guide](/docs/guides/tools-computer-use) for more information.6000 [computer use guide](/api/docs/guides/tools-computer-use) for more information.

5929 6001 

5930 - `id: string`6002 - `id: string`

5931 6003 

6052 A description of the chain of thought used by a reasoning model while generating6124 A description of the chain of thought used by a reasoning model while generating

6053 a response. Be sure to include these items in your `input` to the Responses API6125 a response. Be sure to include these items in your `input` to the Responses API

6054 for subsequent turns of a conversation if you are manually6126 for subsequent turns of a conversation if you are manually

6055 [managing context](/docs/guides/conversation-state).6127 [managing context](/api/docs/guides/conversation-state).

6056 6128 

6057 - `id: string`6129 - `id: string`

6058 6130 

6275 6347 

6276 - `Function object { name, parameters, strict, 6 more }`6348 - `Function object { name, parameters, strict, 6 more }`

6277 6349 

6278 Defines a function in your own code the model can choose to call. Learn more about [function calling](https://platform.openai.com/docs/guides/function-calling).6350 Defines a function in your own code the model can choose to call. Learn more about [function calling](/api/docs/guides/function-calling).

6279 6351 

6280 - `name: string`6352 - `name: string`

6281 6353 

6319 6391 

6320 - `FileSearch object { type, vector_store_ids, filters, 2 more }`6392 - `FileSearch object { type, vector_store_ids, filters, 2 more }`

6321 6393 

6322 A tool that searches for relevant content from uploaded files. Learn more about the [file search tool](https://platform.openai.com/docs/guides/tools-file-search).6394 A tool that searches for relevant content from uploaded files. Learn more about the [file search tool](/api/docs/guides/tools-file-search).

6323 6395 

6324 - `type: "file_search"`6396 - `type: "file_search"`

6325 6397 

6493 6565 

6494 - `Computer object { type }`6566 - `Computer object { type }`

6495 6567 

6496 A tool that controls a virtual computer. Learn more about the [computer tool](https://platform.openai.com/docs/guides/tools-computer-use).6568 A tool that controls a virtual computer. Learn more about the [computer tool](/api/docs/guides/tools-computer-use).

6497 6569 

6498 - `type: "computer"`6570 - `type: "computer"`

6499 6571 

6503 6575 

6504 - `ComputerUsePreview object { display_height, display_width, environment, type }`6576 - `ComputerUsePreview object { display_height, display_width, environment, type }`

6505 6577 

6506 A tool that controls a virtual computer. Learn more about the [computer tool](https://platform.openai.com/docs/guides/tools-computer-use).6578 A tool that controls a virtual computer. Learn more about the [computer tool](/api/docs/guides/tools-computer-use).

6507 6579 

6508 - `display_height: number`6580 - `display_height: number`

6509 6581 

6536 - `WebSearch object { type, external_web_access, filters, 2 more }`6608 - `WebSearch object { type, external_web_access, filters, 2 more }`

6537 6609 

6538 Search the Internet for sources related to the prompt. Learn more about the6610 Search the Internet for sources related to the prompt. Learn more about the

6539 [web search tool](/docs/guides/tools-web-search).6611 [web search tool](/api/docs/guides/tools-web-search).

6540 6612 

6541 - `type: "web_search" or "web_search_2025_08_26"`6613 - `type: "web_search" or "web_search_2025_08_26"`

6542 6614 

6600 - `Mcp object { server_label, type, allowed_callers, 9 more }`6672 - `Mcp object { server_label, type, allowed_callers, 9 more }`

6601 6673 

6602 Give the model access to additional tools via remote Model Context Protocol6674 Give the model access to additional tools via remote Model Context Protocol

6603 (MCP) servers. [Learn more about MCP](/docs/guides/tools-remote-mcp).6675 (MCP) servers. [Learn more about MCP](/api/docs/guides/tools-connectors-mcp).

6604 6676 

6605 - `server_label: string`6677 - `server_label: string`

6606 6678 

6652 6724 

6653 Identifier for service connectors, like those available in ChatGPT. One of6725 Identifier for service connectors, like those available in ChatGPT. One of

6654 `server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more6726 `server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more

6655 about service connectors [here](/docs/guides/tools-remote-mcp#connectors).6727 about service connectors [here](/api/docs/guides/tools-connectors-mcp#connectors).

6656 6728 

6657 Currently supported `connector_id` values are:6729 Currently supported `connector_id` values are:

6658 6730 

6844 6916 

6845 - `background: optional "transparent" or "opaque" or "auto"`6917 - `background: optional "transparent" or "opaque" or "auto"`

6846 6918 

6847 Set the background of the generated image. One of `transparent`,6919 Set the background of the generated image. One of `transparent`, `opaque`,

6848 `opaque`, or `auto`. Transparent backgrounds are available for6920 or `auto`. `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare`, including

6849 supported GPT Image models. For `gpt-image-2` and6921 their `2026-09-08` snapshots, support `opaque` and `transparent`

6850 `gpt-image-2-2026-04-21`, this support is in preview. When using6922 backgrounds. Transparent backgrounds are available for supported GPT Image

6851 `transparent`, set the output format to `png` or `webp`. Default: `auto`.6923 models. For `gpt-image-2` and `gpt-image-2-2026-04-21`, this support is in

6924 preview. When using `transparent`, set the output format to `png` or `webp`.

6925 Default: `auto`.

6852 6926 

6853 - `"transparent"`6927 - `"transparent"`

6854 6928 

6877 6951 

6878 Base64-encoded mask image.6952 Base64-encoded mask image.

6879 6953 

6880 - `model: optional string or "gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 2 more`6954 - `model: optional string or "gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 6 more`

6881 6955 

6882 The image generation model to use. One of `gpt-image-1`,6956 The image generation model to use. One of `gpt-image-1`,

6883 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,6957 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,

6884 `gpt-image-2-2026-04-21`, or `chatgpt-image-latest`. Default:6958 `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`,

6959 `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`,

6960 `gpt-image-2.5-flare-2026-09-08`, or `chatgpt-image-latest`. Default:

6885 `gpt-image-1`.6961 `gpt-image-1`.

6886 6962 

6887 - `string`6963 - `string`

6888 6964 

6889 - `"gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 2 more`6965 - `"gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 6 more`

6890 6966 

6891 The image generation model to use. One of `gpt-image-1`,6967 The image generation model to use. One of `gpt-image-1`,

6892 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,6968 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,

6893 `gpt-image-2-2026-04-21`, or `chatgpt-image-latest`. Default:6969 `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`,

6970 `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`,

6971 `gpt-image-2.5-flare-2026-09-08`, or `chatgpt-image-latest`. Default:

6894 `gpt-image-1`.6972 `gpt-image-1`.

6895 6973 

6896 - `"gpt-image-1"`6974 - `"gpt-image-1"`

6903 6981 

6904 - `"gpt-image-2-2026-04-21"`6982 - `"gpt-image-2-2026-04-21"`

6905 6983 

6984 - `"gpt-image-2.5-sunburst"`

6985 

6986 - `"gpt-image-2.5-sunburst-2026-09-08"`

6987 

6988 - `"gpt-image-2.5-flare"`

6989 

6990 - `"gpt-image-2.5-flare-2026-09-08"`

6991 

6906 - `moderation: optional "auto" or "low"`6992 - `moderation: optional "auto" or "low"`

6907 6993 

6908 Moderation level for the generated image. Default: `auto`.6994 Moderation level for the generated image. Default: `auto`.

6930 7016 

6931 Number of partial images to generate in streaming mode, from 0 (default value) to 3.7017 Number of partial images to generate in streaming mode, from 0 (default value) to 3.

6932 7018 

6933 - `quality: optional "low" or "medium" or "high" or "auto"`7019 - `quality: optional "low" or "medium" or "high" or 3 more`

6934 7020 

6935 The quality of the generated image. One of `low`, `medium`, `high`,7021 The quality of the generated image. The GPT image models support `low`,

6936 or `auto`. Default: `auto`.7022 `medium`, and `high`. `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare`,

7023 including their `2026-09-08` snapshots, also support `xhigh` and `max`.

7024 Default: `auto`.

6937 7025 

6938 - `"low"`7026 - `"low"`

6939 7027 

6941 7029 

6942 - `"high"`7030 - `"high"`

6943 7031 

7032 - `"xhigh"`

7033 

7034 - `"max"`

7035 

6944 - `"auto"`7036 - `"auto"`

6945 7037 

6946 - `size: optional string or "1024x1024" or "1024x1536" or "1536x1024" or "auto"`7038 - `size: optional string or "1024x1024" or "1024x1536" or "1536x1024" or "auto"`

6947 7039 

6948 The size of the generated images. For `gpt-image-2` and `gpt-image-2-2026-04-21`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.7040 The size of the generated images. For `gpt-image-2`, `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`, `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`, and `gpt-image-2.5-flare-2026-09-08`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.

6949 7041 

6950 - `string`7042 - `string`

6951 7043 

6952 - `"1024x1024" or "1024x1536" or "1536x1024" or "auto"`7044 - `"1024x1024" or "1024x1536" or "1536x1024" or "auto"`

6953 7045 

6954 The size of the generated images. For `gpt-image-2` and `gpt-image-2-2026-04-21`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.7046 The size of the generated images. For `gpt-image-2`, `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`, `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`, and `gpt-image-2.5-flare-2026-09-08`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.

6955 7047 

6956 - `"1024x1024"`7048 - `"1024x1024"`

6957 7049 

6999 7091 

7000 - `Custom object { name, type, allowed_callers, 4 more }`7092 - `Custom object { name, type, allowed_callers, 4 more }`

7001 7093 

7002 A custom tool that processes input using a specified format. Learn more about [custom tools](/docs/guides/function-calling#custom-tools)7094 A custom tool that processes input using a specified format. Learn more about [custom tools](/api/docs/guides/function-calling#custom-tools)

7003 7095 

7004 - `name: string`7096 - `name: string`

7005 7097 

7121 7213 

7122 - `Custom object { name, type, allowed_callers, 4 more }`7214 - `Custom object { name, type, allowed_callers, 4 more }`

7123 7215 

7124 A custom tool that processes input using a specified format. Learn more about [custom tools](/docs/guides/function-calling#custom-tools)7216 A custom tool that processes input using a specified format. Learn more about [custom tools](/api/docs/guides/function-calling#custom-tools)

7125 7217 

7126 - `name: string`7218 - `name: string`

7127 7219 

7223 7315 

7224 - `WebSearchPreview object { type, search_content_types, search_context_size, user_location }`7316 - `WebSearchPreview object { type, search_content_types, search_context_size, user_location }`

7225 7317 

7226 This tool searches the web for relevant results to use in a response. Learn more about the [web search tool](https://platform.openai.com/docs/guides/tools-web-search).7318 This tool searches the web for relevant results to use in a response. Learn more about the [web search tool](/api/docs/guides/tools-web-search).

7227 7319 

7228 - `type: "web_search_preview" or "web_search_preview_2025_03_11"`7320 - `type: "web_search_preview" or "web_search_preview_2025_03_11"`

7229 7321 

7343 7435 

7344 - `Function object { name, parameters, strict, 6 more }`7436 - `Function object { name, parameters, strict, 6 more }`

7345 7437 

7346 Defines a function in your own code the model can choose to call. Learn more about [function calling](https://platform.openai.com/docs/guides/function-calling).7438 Defines a function in your own code the model can choose to call. Learn more about [function calling](/api/docs/guides/function-calling).

7347 7439 

7348 - `name: string`7440 - `name: string`

7349 7441 

7387 7479 

7388 - `FileSearch object { type, vector_store_ids, filters, 2 more }`7480 - `FileSearch object { type, vector_store_ids, filters, 2 more }`

7389 7481 

7390 A tool that searches for relevant content from uploaded files. Learn more about the [file search tool](https://platform.openai.com/docs/guides/tools-file-search).7482 A tool that searches for relevant content from uploaded files. Learn more about the [file search tool](/api/docs/guides/tools-file-search).

7391 7483 

7392 - `type: "file_search"`7484 - `type: "file_search"`

7393 7485 

7561 7653 

7562 - `Computer object { type }`7654 - `Computer object { type }`

7563 7655 

7564 A tool that controls a virtual computer. Learn more about the [computer tool](https://platform.openai.com/docs/guides/tools-computer-use).7656 A tool that controls a virtual computer. Learn more about the [computer tool](/api/docs/guides/tools-computer-use).

7565 7657 

7566 - `type: "computer"`7658 - `type: "computer"`

7567 7659 

7571 7663 

7572 - `ComputerUsePreview object { display_height, display_width, environment, type }`7664 - `ComputerUsePreview object { display_height, display_width, environment, type }`

7573 7665 

7574 A tool that controls a virtual computer. Learn more about the [computer tool](https://platform.openai.com/docs/guides/tools-computer-use).7666 A tool that controls a virtual computer. Learn more about the [computer tool](/api/docs/guides/tools-computer-use).

7575 7667 

7576 - `display_height: number`7668 - `display_height: number`

7577 7669 

7604 - `WebSearch object { type, external_web_access, filters, 2 more }`7696 - `WebSearch object { type, external_web_access, filters, 2 more }`

7605 7697 

7606 Search the Internet for sources related to the prompt. Learn more about the7698 Search the Internet for sources related to the prompt. Learn more about the

7607 [web search tool](/docs/guides/tools-web-search).7699 [web search tool](/api/docs/guides/tools-web-search).

7608 7700 

7609 - `type: "web_search" or "web_search_2025_08_26"`7701 - `type: "web_search" or "web_search_2025_08_26"`

7610 7702 

7668 - `Mcp object { server_label, type, allowed_callers, 9 more }`7760 - `Mcp object { server_label, type, allowed_callers, 9 more }`

7669 7761 

7670 Give the model access to additional tools via remote Model Context Protocol7762 Give the model access to additional tools via remote Model Context Protocol

7671 (MCP) servers. [Learn more about MCP](/docs/guides/tools-remote-mcp).7763 (MCP) servers. [Learn more about MCP](/api/docs/guides/tools-connectors-mcp).

7672 7764 

7673 - `server_label: string`7765 - `server_label: string`

7674 7766 

7720 7812 

7721 Identifier for service connectors, like those available in ChatGPT. One of7813 Identifier for service connectors, like those available in ChatGPT. One of

7722 `server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more7814 `server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more

7723 about service connectors [here](/docs/guides/tools-remote-mcp#connectors).7815 about service connectors [here](/api/docs/guides/tools-connectors-mcp#connectors).

7724 7816 

7725 Currently supported `connector_id` values are:7817 Currently supported `connector_id` values are:

7726 7818 

7912 8004 

7913 - `background: optional "transparent" or "opaque" or "auto"`8005 - `background: optional "transparent" or "opaque" or "auto"`

7914 8006 

7915 Set the background of the generated image. One of `transparent`,8007 Set the background of the generated image. One of `transparent`, `opaque`,

7916 `opaque`, or `auto`. Transparent backgrounds are available for8008 or `auto`. `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare`, including

7917 supported GPT Image models. For `gpt-image-2` and8009 their `2026-09-08` snapshots, support `opaque` and `transparent`

7918 `gpt-image-2-2026-04-21`, this support is in preview. When using8010 backgrounds. Transparent backgrounds are available for supported GPT Image

7919 `transparent`, set the output format to `png` or `webp`. Default: `auto`.8011 models. For `gpt-image-2` and `gpt-image-2-2026-04-21`, this support is in

8012 preview. When using `transparent`, set the output format to `png` or `webp`.

8013 Default: `auto`.

7920 8014 

7921 - `"transparent"`8015 - `"transparent"`

7922 8016 

7945 8039 

7946 Base64-encoded mask image.8040 Base64-encoded mask image.

7947 8041 

7948 - `model: optional string or "gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 2 more`8042 - `model: optional string or "gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 6 more`

7949 8043 

7950 The image generation model to use. One of `gpt-image-1`,8044 The image generation model to use. One of `gpt-image-1`,

7951 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,8045 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,

7952 `gpt-image-2-2026-04-21`, or `chatgpt-image-latest`. Default:8046 `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`,

8047 `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`,

8048 `gpt-image-2.5-flare-2026-09-08`, or `chatgpt-image-latest`. Default:

7953 `gpt-image-1`.8049 `gpt-image-1`.

7954 8050 

7955 - `string`8051 - `string`

7956 8052 

7957 - `"gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 2 more`8053 - `"gpt-image-1" or "gpt-image-1-mini" or "gpt-image-1.5" or 6 more`

7958 8054 

7959 The image generation model to use. One of `gpt-image-1`,8055 The image generation model to use. One of `gpt-image-1`,

7960 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,8056 `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,

7961 `gpt-image-2-2026-04-21`, or `chatgpt-image-latest`. Default:8057 `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`,

8058 `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`,

8059 `gpt-image-2.5-flare-2026-09-08`, or `chatgpt-image-latest`. Default:

7962 `gpt-image-1`.8060 `gpt-image-1`.

7963 8061 

7964 - `"gpt-image-1"`8062 - `"gpt-image-1"`

7971 8069 

7972 - `"gpt-image-2-2026-04-21"`8070 - `"gpt-image-2-2026-04-21"`

7973 8071 

8072 - `"gpt-image-2.5-sunburst"`

8073 

8074 - `"gpt-image-2.5-sunburst-2026-09-08"`

8075 

8076 - `"gpt-image-2.5-flare"`

8077 

8078 - `"gpt-image-2.5-flare-2026-09-08"`

8079 

7974 - `moderation: optional "auto" or "low"`8080 - `moderation: optional "auto" or "low"`

7975 8081 

7976 Moderation level for the generated image. Default: `auto`.8082 Moderation level for the generated image. Default: `auto`.

7998 8104 

7999 Number of partial images to generate in streaming mode, from 0 (default value) to 3.8105 Number of partial images to generate in streaming mode, from 0 (default value) to 3.

8000 8106 

8001 - `quality: optional "low" or "medium" or "high" or "auto"`8107 - `quality: optional "low" or "medium" or "high" or 3 more`

8002 8108 

8003 The quality of the generated image. One of `low`, `medium`, `high`,8109 The quality of the generated image. The GPT image models support `low`,

8004 or `auto`. Default: `auto`.8110 `medium`, and `high`. `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare`,

8111 including their `2026-09-08` snapshots, also support `xhigh` and `max`.

8112 Default: `auto`.

8005 8113 

8006 - `"low"`8114 - `"low"`

8007 8115 

8009 8117 

8010 - `"high"`8118 - `"high"`

8011 8119 

8120 - `"xhigh"`

8121 

8122 - `"max"`

8123 

8012 - `"auto"`8124 - `"auto"`

8013 8125 

8014 - `size: optional string or "1024x1024" or "1024x1536" or "1536x1024" or "auto"`8126 - `size: optional string or "1024x1024" or "1024x1536" or "1536x1024" or "auto"`

8015 8127 

8016 The size of the generated images. For `gpt-image-2` and `gpt-image-2-2026-04-21`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.8128 The size of the generated images. For `gpt-image-2`, `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`, `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`, and `gpt-image-2.5-flare-2026-09-08`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.

8017 8129 

8018 - `string`8130 - `string`

8019 8131 

8020 - `"1024x1024" or "1024x1536" or "1536x1024" or "auto"`8132 - `"1024x1024" or "1024x1536" or "1536x1024" or "auto"`

8021 8133 

8022 The size of the generated images. For `gpt-image-2` and `gpt-image-2-2026-04-21`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.8134 The size of the generated images. For `gpt-image-2`, `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`, `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`, and `gpt-image-2.5-flare-2026-09-08`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.

8023 8135