Reference 2026-07-29 15:02 UTC to 2026-07-30 23:58 UTC

1125 1125 

1126 - `"incomplete"`1126 - `"incomplete"`

1127 1127 

1128 - `FunctionCallOutput object { call_id, output, type, 4 more }`1128 - `FunctionCallOutput object { call_id, output, type, 6 more }`

1129 1129 

1130 The output of a function tool call.1130 The output of a function tool call.

1131 1131 

1297 1297 

1298 - `"program"`1298 - `"program"`

1299 1299 

1300 - `name: optional string`

1301 

1302 The name of the tool that produced the output.

1303 

1304 - `namespace: optional string`

1305 

1306 The namespace of the tool that produced the output.

1307 

1300 - `status: optional "in_progress" or "completed" or "incomplete"`1308 - `status: optional "in_progress" or "completed" or "incomplete"`

1301 1309 

1302 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.1310 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

5264 5272 

5265 - `"incomplete"`5273 - `"incomplete"`

5266 5274 

5267 - `FunctionCallOutput object { id, call_id, output, 5 more }`5275 - `FunctionCallOutput object { id, call_id, output, 7 more }`

5268 5276 

5269 - `id: string`5277 - `id: string`

5270 5278 

5352 5360 

5353 The identifier of the actor that created the item.5361 The identifier of the actor that created the item.

5354 5362 

5363 - `name: optional string`

5364 

5365 The name of the tool that produced the output.

5366 

5367 - `namespace: optional string`

5368 

5369 The namespace of the tool that produced the output.

5370 

5355 - `AgentMessage object { id, author, content, 3 more }`5371 - `AgentMessage object { id, author, content, 3 more }`

5356 5372 

5357 - `id: string`5373 - `id: string`

10564 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.10580 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

10565 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).10581 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).

10566 10582 

10567 - `service_tier: optional "auto" or "default" or "flex" or 2 more`10583 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

10568 10584 

10569 Specifies the processing type used for serving the request.10585 Specifies the processing type used for serving the request.

10570 10586 

10571 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.10587 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

10572 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.10588 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

10573 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.10589 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

10590 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

10574 - When not set, the default behavior is 'auto'.10591 - When not set, the default behavior is 'auto'.

10575 10592 

10576 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.10593 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

10585 10602 

10586 - `"priority"`10603 - `"priority"`

10587 10604 

10605 - `"fast"`

10606 

10588 - `status: optional BetaResponseStatus`10607 - `status: optional BetaResponseStatus`

10589 10608 

10590 The status of the response generation. One of `completed`, `failed`,10609 The status of the response generation. One of `completed`, `failed`,

1241 1241 

1242 - `"incomplete"`1242 - `"incomplete"`

1243 1243 

1244 - `FunctionCallOutput object { call_id, output, type, 4 more }`1244 - `FunctionCallOutput object { call_id, output, type, 6 more }`

1245 1245 

1246 The output of a function tool call.1246 The output of a function tool call.

1247 1247 

1413 1413 

1414 - `"program"`1414 - `"program"`

1415 1415 

1416 - `name: optional string`

1417 

1418 The name of the tool that produced the output.

1419 

1420 - `namespace: optional string`

1421 

1422 The namespace of the tool that produced the output.

1423 

1416 - `status: optional "in_progress" or "completed" or "incomplete"`1424 - `status: optional "in_progress" or "completed" or "incomplete"`

1417 1425 

1418 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.1426 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

5037 5045 

5038 - `"24h"`5046 - `"24h"`

5039 5047 

5040- `service_tier: optional "auto" or "default" or "flex" or "priority"`5048- `service_tier: optional "auto" or "default" or "fast" or 2 more`

5041 5049 

5042 The service tier to use for this request.5050 Specifies the processing type used for serving the request. - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'. - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model. - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier. - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request. - When not set, the default behavior is 'auto'.

5051 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

5043 5052 

5044 - `"auto"`5053 - `"auto"`

5045 5054 

5046 - `"default"`5055 - `"default"`

5047 5056 

5057 - `"fast"`

5058 

5048 - `"flex"`5059 - `"flex"`

5049 5060 

5050 - `"priority"`5061 - `"priority"`

8137 8148 

8138 The canonical name of the agent that produced this item.8149 The canonical name of the agent that produced this item.

8139 8150 

8140 - `FunctionCallOutput object { call_id, output, type, 4 more }`8151 - `FunctionCallOutput object { call_id, output, type, 6 more }`

8141 8152 

8142 The output of a function tool call.8153 The output of a function tool call.

8143 8154 

8213 8224 

8214 - `"program"`8225 - `"program"`

8215 8226 

8227 - `name: optional string`

8228 

8229 The name of the tool that produced the output.

8230 

8231 - `namespace: optional string`

8232 

8233 The namespace of the tool that produced the output.

8234 

8216 - `status: optional "in_progress" or "completed" or "incomplete"`8235 - `status: optional "in_progress" or "completed" or "incomplete"`

8217 8236 

8218 The status of the item. One of `in_progress`, `completed`, or8237 The status of the item. One of `in_progress`, `completed`, or

1117 1117 

1118 - `"incomplete"`1118 - `"incomplete"`

1119 1119 

1120 - `FunctionCallOutput object { call_id, output, type, 4 more }`1120 - `FunctionCallOutput object { call_id, output, type, 6 more }`

1121 1121 

1122 The output of a function tool call.1122 The output of a function tool call.

1123 1123 

1289 1289 

1290 - `"program"`1290 - `"program"`

1291 1291 

1292 - `name: optional string`

1293 

1294 The name of the tool that produced the output.

1295 

1296 - `namespace: optional string`

1297 

1298 The namespace of the tool that produced the output.

1299 

1292 - `status: optional "in_progress" or "completed" or "incomplete"`1300 - `status: optional "in_progress" or "completed" or "incomplete"`

1293 1301 

1294 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.1302 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

5332 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.5340 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

5333 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).5341 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).

5334 5342 

5335- `service_tier: optional "auto" or "default" or "flex" or 2 more`5343- `service_tier: optional "auto" or "default" or "flex" or 3 more`

5336 5344 

5337 Specifies the processing type used for serving the request.5345 Specifies the processing type used for serving the request.

5338 5346 

5339 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.5347 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

5340 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.5348 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

5341 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.5349 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

5350 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

5342 - When not set, the default behavior is 'auto'.5351 - When not set, the default behavior is 'auto'.

5343 5352 

5344 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.5353 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

5353 5362 

5354 - `"priority"`5363 - `"priority"`

5355 5364 

5365 - `"fast"`

5366 

5356- `store: optional boolean`5367- `store: optional boolean`

5357 5368 

5358 Whether to store the generated model response for later retrieval via5369 Whether to store the generated model response for later retrieval via

7802 7813 

7803 - `"incomplete"`7814 - `"incomplete"`

7804 7815 

7805 - `FunctionCallOutput object { call_id, output, type, 4 more }`7816 - `FunctionCallOutput object { call_id, output, type, 6 more }`

7806 7817 

7807 The output of a function tool call.7818 The output of a function tool call.

7808 7819 

7974 7985 

7975 - `"program"`7986 - `"program"`

7976 7987 

7988 - `name: optional string`

7989 

7990 The name of the tool that produced the output.

7991 

7992 - `namespace: optional string`

7993 

7994 The namespace of the tool that produced the output.

7995 

7977 - `status: optional "in_progress" or "completed" or "incomplete"`7996 - `status: optional "in_progress" or "completed" or "incomplete"`

7978 7997 

7979 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.7998 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

11941 11960 

11942 - `"incomplete"`11961 - `"incomplete"`

11943 11962 

11944 - `FunctionCallOutput object { id, call_id, output, 5 more }`11963 - `FunctionCallOutput object { id, call_id, output, 7 more }`

11945 11964 

11946 - `id: string`11965 - `id: string`

11947 11966 

12029 12048 

12030 The identifier of the actor that created the item.12049 The identifier of the actor that created the item.

12031 12050 

12051 - `name: optional string`

12052 

12053 The name of the tool that produced the output.

12054 

12055 - `namespace: optional string`

12056 

12057 The namespace of the tool that produced the output.

12058 

12032 - `AgentMessage object { id, author, content, 3 more }`12059 - `AgentMessage object { id, author, content, 3 more }`

12033 12060 

12034 - `id: string`12061 - `id: string`

17241 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.17268 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

17242 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).17269 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).

17243 17270 

17244 - `service_tier: optional "auto" or "default" or "flex" or 2 more`17271 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

17245 17272 

17246 Specifies the processing type used for serving the request.17273 Specifies the processing type used for serving the request.

17247 17274 

17248 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.17275 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

17249 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.17276 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

17250 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.17277 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

17278 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

17251 - When not set, the default behavior is 'auto'.17279 - When not set, the default behavior is 'auto'.

17252 17280 

17253 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.17281 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

17262 17290 

17263 - `"priority"`17291 - `"priority"`

17264 17292 

17293 - `"fast"`

17294 

17265 - `status: optional BetaResponseStatus`17295 - `status: optional BetaResponseStatus`

17266 17296 

17267 The status of the response generation. One of `completed`, `failed`,17297 The status of the response generation. One of `completed`, `failed`,

1169 1169 

1170 - `"incomplete"`1170 - `"incomplete"`

1171 1171 

1172 - `FunctionCallOutput object { call_id, output, type, 4 more }`1172 - `FunctionCallOutput object { call_id, output, type, 6 more }`

1173 1173 

1174 The output of a function tool call.1174 The output of a function tool call.

1175 1175 

1341 1341 

1342 - `"program"`1342 - `"program"`

1343 1343 

1344 - `name: optional string`

1345 

1346 The name of the tool that produced the output.

1347 

1348 - `namespace: optional string`

1349 

1350 The namespace of the tool that produced the output.

1351 

1344 - `status: optional "in_progress" or "completed" or "incomplete"`1352 - `status: optional "in_progress" or "completed" or "incomplete"`

1345 1353 

1346 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.1354 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

5308 5316 

5309 - `"incomplete"`5317 - `"incomplete"`

5310 5318 

5311 - `FunctionCallOutput object { id, call_id, output, 5 more }`5319 - `FunctionCallOutput object { id, call_id, output, 7 more }`

5312 5320 

5313 - `id: string`5321 - `id: string`

5314 5322 

5396 5404 

5397 The identifier of the actor that created the item.5405 The identifier of the actor that created the item.

5398 5406 

5407 - `name: optional string`

5408 

5409 The name of the tool that produced the output.

5410 

5411 - `namespace: optional string`

5412 

5413 The namespace of the tool that produced the output.

5414 

5399 - `AgentMessage object { id, author, content, 3 more }`5415 - `AgentMessage object { id, author, content, 3 more }`

5400 5416 

5401 - `id: string`5417 - `id: string`

10608 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.10624 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

10609 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).10625 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).

10610 10626 

10611 - `service_tier: optional "auto" or "default" or "flex" or 2 more`10627 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

10612 10628 

10613 Specifies the processing type used for serving the request.10629 Specifies the processing type used for serving the request.

10614 10630 

10615 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.10631 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

10616 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.10632 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

10617 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.10633 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

10634 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

10618 - When not set, the default behavior is 'auto'.10635 - When not set, the default behavior is 'auto'.

10619 10636 

10620 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.10637 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

10629 10646 

10630 - `"priority"`10647 - `"priority"`

10631 10648 

10649 - `"fast"`

10650 

10632 - `status: optional BetaResponseStatus`10651 - `status: optional BetaResponseStatus`

10633 10652 

10634 The status of the response generation. One of `completed`, `failed`,10653 The status of the response generation. One of `completed`, `failed`,

1038 1038 

1039 The namespace of the function to run.1039 The namespace of the function to run.

1040 1040 

1041 - `FunctionCallOutput object { id, call_id, output, 5 more }`1041 - `FunctionCallOutput object { id, call_id, output, 7 more }`

1042 1042 

1043 - `id: string`1043 - `id: string`

1044 1044 

1126 1126 

1127 The identifier of the actor that created the item.1127 The identifier of the actor that created the item.

1128 1128 

1129 - `name: optional string`

1130 

1131 The name of the tool that produced the output.

1132 

1133 - `namespace: optional string`

1134 

1135 The namespace of the tool that produced the output.

1136 

1129 - `AgentMessage object { id, author, content, 3 more }`1137 - `AgentMessage object { id, author, content, 3 more }`

1130 1138 

1131 - `id: string`1139 - `id: string`

1058 1058 

1059 - `"incomplete"`1059 - `"incomplete"`

1060 1060 

1061 - `FunctionCallOutput object { call_id, output, type, 4 more }`1061 - `FunctionCallOutput object { call_id, output, type, 6 more }`

1062 1062 

1063 The output of a function tool call.1063 The output of a function tool call.

1064 1064 

1230 1230 

1231 - `"program"`1231 - `"program"`

1232 1232 

1233 - `name: optional string`

1234 

1235 The name of the tool that produced the output.

1236 

1237 - `namespace: optional string`

1238 

1239 The namespace of the tool that produced the output.

1240 

1233 - `status: optional "in_progress" or "completed" or "incomplete"`1241 - `status: optional "in_progress" or "completed" or "incomplete"`

1234 1242 

1235 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.1243 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

1032 If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same `seed` and parameters should return the same result.1032 If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same `seed` and parameters should return the same result.

1033 Determinism is not guaranteed, and you should refer to the `system_fingerprint` response parameter to monitor changes in the backend.1033 Determinism is not guaranteed, and you should refer to the `system_fingerprint` response parameter to monitor changes in the backend.

1034 1034 

1035- `service_tier: optional "auto" or "default" or "flex" or 2 more`1035- `service_tier: optional "auto" or "default" or "flex" or 3 more`

1036 1036 

1037 Specifies the processing type used for serving the request.1037 Specifies the processing type used for serving the request.

1038 1038 

1039 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.1039 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

1040 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.1040 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

1041 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.1041 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

1042 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

1042 - When not set, the default behavior is 'auto'.1043 - When not set, the default behavior is 'auto'.

1043 1044 

1044 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.1045 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

1053 1054 

1054 - `"priority"`1055 - `"priority"`

1055 1056 

1057 - `"fast"`

1058 

1056- `stop: optional string or array of string`1059- `stop: optional string or array of string`

1057 1060 

1058 Not supported with latest reasoning models `o3` and `o4-mini`.1061 Not supported with latest reasoning models `o3` and `o4-mini`.

1777 1780 

1778 - `"error"`1781 - `"error"`

1779 1782 

1780 - `service_tier: optional "auto" or "default" or "flex" or 2 more`1783 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

1781 1784 

1782 Specifies the processing type used for serving the request.1785 Specifies the processing type used for serving the request.

1783 1786 

1784 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.1787 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

1785 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.1788 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

1786 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.1789 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

1790 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

1787 - When not set, the default behavior is 'auto'.1791 - When not set, the default behavior is 'auto'.

1788 1792 

1789 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.1793 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

1798 1802 

1799 - `"priority"`1803 - `"priority"`

1800 1804 

1805 - `"fast"`

1806 

1801 - `system_fingerprint: optional string`1807 - `system_fingerprint: optional string`

1802 1808 

1803 This fingerprint represents the backend configuration that the model runs with.1809 This fingerprint represents the backend configuration that the model runs with.

425 425 

426 - `"error"`426 - `"error"`

427 427 

428 - `service_tier: optional "auto" or "default" or "flex" or 2 more`428 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

429 429 

430 Specifies the processing type used for serving the request.430 Specifies the processing type used for serving the request.

431 431 

432 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.432 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

433 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.433 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

434 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.434 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

435 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

435 - When not set, the default behavior is 'auto'.436 - When not set, the default behavior is 'auto'.

436 437 

437 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.438 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

446 447 

447 - `"priority"`448 - `"priority"`

448 449 

450 - `"fast"`

451 

449 - `system_fingerprint: optional string`452 - `system_fingerprint: optional string`

450 453 

451 This fingerprint represents the backend configuration that the model runs with.454 This fingerprint represents the backend configuration that the model runs with.

401 401 

402 - `"error"`402 - `"error"`

403 403 

404 - `service_tier: optional "auto" or "default" or "flex" or 2 more`404 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

405 405 

406 Specifies the processing type used for serving the request.406 Specifies the processing type used for serving the request.

407 407 

408 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.408 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

409 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.409 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

410 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.410 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

411 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

411 - When not set, the default behavior is 'auto'.412 - When not set, the default behavior is 'auto'.

412 413 

413 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.414 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

422 423 

423 - `"priority"`424 - `"priority"`

424 425 

426 - `"fast"`

427 

425 - `system_fingerprint: optional string`428 - `system_fingerprint: optional string`

426 429 

427 This fingerprint represents the backend configuration that the model runs with.430 This fingerprint represents the backend configuration that the model runs with.

413 413 

414 - `"error"`414 - `"error"`

415 415 

416 - `service_tier: optional "auto" or "default" or "flex" or 2 more`416 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

417 417 

418 Specifies the processing type used for serving the request.418 Specifies the processing type used for serving the request.

419 419 

420 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.420 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

421 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.421 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

422 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.422 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

423 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

423 - When not set, the default behavior is 'auto'.424 - When not set, the default behavior is 'auto'.

424 425 

425 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.426 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

434 435 

435 - `"priority"`436 - `"priority"`

436 437 

438 - `"fast"`

439 

437 - `system_fingerprint: optional string`440 - `system_fingerprint: optional string`

438 441 

439 This fingerprint represents the backend configuration that the model runs with.442 This fingerprint represents the backend configuration that the model runs with.

206 "oasRef": "#/components/schemas/CreateChatCompletionStreamResponse/properties/service_tier",206 "oasRef": "#/components/schemas/CreateChatCompletionStreamResponse/properties/service_tier",

207 "deprecated": false,207 "deprecated": false,

208 "key": "service_tier",208 "key": "service_tier",

209 "docstring": "Specifies the processing type used for serving the request.\n - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.\n - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.\n - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.\n - When not set, the default behavior is 'auto'.\n\n When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.\n",209 "docstring": "Specifies the processing type used for serving the request.\n - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.\n - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.\n - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.\n - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.\n - When not set, the default behavior is 'auto'.\n\n When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.\n",

210 "type": {210 "type": {

211 "kind": "HttpTypeUnion",211 "kind": "HttpTypeUnion",

212 "oasRef": "#/components/schemas/CreateChatCompletionStreamResponse/properties/service_tier",212 "oasRef": "#/components/schemas/CreateChatCompletionStreamResponse/properties/service_tier",

230 {230 {

231 "kind": "HttpTypeLiteral",231 "kind": "HttpTypeLiteral",

232 "literal": "priority"232 "literal": "priority"

233 },

234 {

235 "kind": "HttpTypeLiteral",

236 "literal": "fast"

233 }237 }

234 ]238 ]

235 },239 },

243 "(resource) chat.completions > (model) chat_completion_chunk > (schema) > (property) service_tier > (member) 1",247 "(resource) chat.completions > (model) chat_completion_chunk > (schema) > (property) service_tier > (member) 1",

244 "(resource) chat.completions > (model) chat_completion_chunk > (schema) > (property) service_tier > (member) 2",248 "(resource) chat.completions > (model) chat_completion_chunk > (schema) > (property) service_tier > (member) 2",

245 "(resource) chat.completions > (model) chat_completion_chunk > (schema) > (property) service_tier > (member) 3",249 "(resource) chat.completions > (model) chat_completion_chunk > (schema) > (property) service_tier > (member) 3",

246 "(resource) chat.completions > (model) chat_completion_chunk > (schema) > (property) service_tier > (member) 4"250 "(resource) chat.completions > (model) chat_completion_chunk > (schema) > (property) service_tier > (member) 4",

251 "(resource) chat.completions > (model) chat_completion_chunk > (schema) > (property) service_tier > (member) 5"

247 ]252 ]

248 },253 },

249 "(resource) chat.completions > (model) chat_completion_chunk > (schema) > (property) system_fingerprint": {254 "(resource) chat.completions > (model) chat_completion_chunk > (schema) > (property) system_fingerprint": {

547 "literal": "priority"552 "literal": "priority"

548 }553 }

549 },554 },

555 "(resource) chat.completions > (model) chat_completion_chunk > (schema) > (property) service_tier > (member) 5": {

556 "kind": "HttpDeclReference",

557 "type": {

558 "kind": "HttpTypeLiteral",

559 "literal": "fast"

560 }

561 },

550 "(resource) completions > (model) completion_usage > (schema) > (property) completion_tokens": {562 "(resource) completions > (model) completion_usage > (schema) > (property) completion_tokens": {

551 "kind": "HttpDeclProperty",563 "kind": "HttpDeclProperty",

552 "oasRef": "#/components/schemas/CompletionUsage/properties/completion_tokens",564 "oasRef": "#/components/schemas/CompletionUsage/properties/completion_tokens",

969 969 

970 - `"incomplete"`970 - `"incomplete"`

971 971 

972 - `FunctionCallOutput object { call_id, output, type, 3 more }`972 - `FunctionCallOutput object { call_id, output, type, 5 more }`

973 973 

974 The output of a function tool call.974 The output of a function tool call.

975 975 

1133 1133 

1134 - `"program"`1134 - `"program"`

1135 1135 

1136 - `name: optional string`

1137 

1138 The name of the tool that produced the output.

1139 

1140 - `namespace: optional string`

1141 

1142 The namespace of the tool that produced the output.

1143 

1136 - `status: optional "in_progress" or "completed" or "incomplete"`1144 - `status: optional "in_progress" or "completed" or "incomplete"`

1137 1145 

1138 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.1146 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

996 996 

997 - `"incomplete"`997 - `"incomplete"`

998 998 

999 - `FunctionCallOutput object { call_id, output, type, 3 more }`999 - `FunctionCallOutput object { call_id, output, type, 5 more }`

1000 1000 

1001 The output of a function tool call.1001 The output of a function tool call.

1002 1002 

1160 1160 

1161 - `"program"`1161 - `"program"`

1162 1162 

1163 - `name: optional string`

1164 

1165 The name of the tool that produced the output.

1166 

1167 - `namespace: optional string`

1168 

1169 The namespace of the tool that produced the output.

1170 

1163 - `status: optional "in_progress" or "completed" or "incomplete"`1171 - `status: optional "in_progress" or "completed" or "incomplete"`

1164 1172 

1165 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.1173 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

4531 4539 

4532 The namespace of the function to run.4540 The namespace of the function to run.

4533 4541 

4534 - `FunctionCallOutput object { id, call_id, output, 4 more }`4542 - `FunctionCallOutput object { id, call_id, output, 6 more }`

4535 4543 

4536 - `id: string`4544 - `id: string`

4537 4545 

4611 4619 

4612 The identifier of the actor that created the item.4620 The identifier of the actor that created the item.

4613 4621 

4622 - `name: optional string`

4623 

4624 The name of the tool that produced the output.

4625 

4626 - `namespace: optional string`

4627 

4628 The namespace of the tool that produced the output.

4629 

4614 - `FileSearchCall object { id, queries, status, 2 more }`4630 - `FileSearchCall object { id, queries, status, 2 more }`

4615 4631 

4616 The results of a file search tool call. See the4632 The results of a file search tool call. See the

519 519 

520 The namespace of the function to run.520 The namespace of the function to run.

521 521 

522 - `FunctionCallOutput object { id, call_id, output, 4 more }`522 - `FunctionCallOutput object { id, call_id, output, 6 more }`

523 523 

524 - `id: string`524 - `id: string`

525 525 

599 599 

600 The identifier of the actor that created the item.600 The identifier of the actor that created the item.

601 601 

602 - `name: optional string`

603 

604 The name of the tool that produced the output.

605 

606 - `namespace: optional string`

607 

608 The namespace of the tool that produced the output.

609 

602 - `FileSearchCall object { id, queries, status, 2 more }`610 - `FileSearchCall object { id, queries, status, 2 more }`

603 611 

604 The results of a file search tool call. See the612 The results of a file search tool call. See the

35 35 

36### Returns36### Returns

37 37 

38- `ConversationItem = Message or object { id, arguments, call_id, 6 more } or object { id, call_id, output, 4 more } or 25 more`38- `ConversationItem = Message or object { id, arguments, call_id, 6 more } or object { id, call_id, output, 6 more } or 25 more`

39 39 

40 A single item within a conversation. The set of possible types are the same as the `output` type of a [Response object](/docs/api-reference/responses/object#responses/object-output).40 A single item within a conversation. The set of possible types are the same as the `output` type of a [Response object](/docs/api-reference/responses/object#responses/object-output).

41 41 

490 490 

491 The namespace of the function to run.491 The namespace of the function to run.

492 492 

493 - `FunctionCallOutput object { id, call_id, output, 4 more }`493 - `FunctionCallOutput object { id, call_id, output, 6 more }`

494 494 

495 - `id: string`495 - `id: string`

496 496 

570 570 

571 The identifier of the actor that created the item.571 The identifier of the actor that created the item.

572 572 

573 - `name: optional string`

574 

575 The name of the tool that produced the output.

576 

577 - `namespace: optional string`

578 

579 The namespace of the tool that produced the output.

580 

573 - `FileSearchCall object { id, queries, status, 2 more }`581 - `FileSearchCall object { id, queries, status, 2 more }`

574 582 

575 The results of a file search tool call. See the583 The results of a file search tool call. See the

1063 1063 

1064 - `"incomplete"`1064 - `"incomplete"`

1065 1065 

1066 - `FunctionCallOutput object { call_id, output, type, 3 more }`1066 - `FunctionCallOutput object { call_id, output, type, 5 more }`

1067 1067 

1068 The output of a function tool call.1068 The output of a function tool call.

1069 1069 

1227 1227 

1228 - `"program"`1228 - `"program"`

1229 1229 

1230 - `name: optional string`

1231 

1232 The name of the tool that produced the output.

1233 

1234 - `namespace: optional string`

1235 

1236 The namespace of the tool that produced the output.

1237 

1230 - `status: optional "in_progress" or "completed" or "incomplete"`1238 - `status: optional "in_progress" or "completed" or "incomplete"`

1231 1239 

1232 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.1240 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

4500 4508 

4501 - `"incomplete"`4509 - `"incomplete"`

4502 4510 

4503 - `FunctionCallOutput object { id, call_id, output, 4 more }`4511 - `FunctionCallOutput object { id, call_id, output, 6 more }`

4504 4512 

4505 - `id: string`4513 - `id: string`

4506 4514 

4580 4588 

4581 The identifier of the actor that created the item.4589 The identifier of the actor that created the item.

4582 4590 

4591 - `name: optional string`

4592 

4593 The name of the tool that produced the output.

4594 

4595 - `namespace: optional string`

4596 

4597 The namespace of the tool that produced the output.

4598 

4583 - `WebSearchCall object { id, action, status, type }`4599 - `WebSearchCall object { id, action, status, type }`

4584 4600 

4585 The results of a web search tool call. See the4601 The results of a web search tool call. See the

8812 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.8828 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

8813 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).8829 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).

8814 8830 

8815 - `service_tier: optional "auto" or "default" or "flex" or 2 more`8831 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

8816 8832 

8817 Specifies the processing type used for serving the request.8833 Specifies the processing type used for serving the request.

8818 8834 

8819 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.8835 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

8820 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.8836 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

8821 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.8837 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

8838 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

8822 - When not set, the default behavior is 'auto'.8839 - When not set, the default behavior is 'auto'.

8823 8840 

8824 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.8841 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

8833 8850 

8834 - `"priority"`8851 - `"priority"`

8835 8852 

8853 - `"fast"`

8854 

8836 - `status: optional ResponseStatus`8855 - `status: optional ResponseStatus`

8837 8856 

8838 The status of the response generation. One of `completed`, `failed`,8857 The status of the response generation. One of `completed`, `failed`,

1179 1179 

1180 - `"incomplete"`1180 - `"incomplete"`

1181 1181 

1182 - `FunctionCallOutput object { call_id, output, type, 3 more }`1182 - `FunctionCallOutput object { call_id, output, type, 5 more }`

1183 1183 

1184 The output of a function tool call.1184 The output of a function tool call.

1185 1185 

1343 1343 

1344 - `"program"`1344 - `"program"`

1345 1345 

1346 - `name: optional string`

1347 

1348 The name of the tool that produced the output.

1349 

1350 - `namespace: optional string`

1351 

1352 The namespace of the tool that produced the output.

1353 

1346 - `status: optional "in_progress" or "completed" or "incomplete"`1354 - `status: optional "in_progress" or "completed" or "incomplete"`

1347 1355 

1348 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.1356 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

4292 4300 

4293 - `"24h"`4301 - `"24h"`

4294 4302 

4295- `service_tier: optional "auto" or "default" or "flex" or "priority"`4303- `service_tier: optional "auto" or "default" or "fast" or 2 more`

4296 4304 

4297 The service tier to use for this request.4305 Specifies the processing type used for serving the request. - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'. - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model. - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier. - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request. - When not set, the default behavior is 'auto'.

4306 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

4298 4307 

4299 - `"auto"`4308 - `"auto"`

4300 4309 

4301 - `"default"`4310 - `"default"`

4302 4311 

4312 - `"fast"`

4313 

4303 - `"flex"`4314 - `"flex"`

4304 4315 

4305 - `"priority"`4316 - `"priority"`

6815 6826 

6816 - `"additional_tools"`6827 - `"additional_tools"`

6817 6828 

6818 - `FunctionCallOutput object { call_id, output, type, 3 more }`6829 - `FunctionCallOutput object { call_id, output, type, 5 more }`

6819 6830 

6820 The output of a function tool call.6831 The output of a function tool call.

6821 6832 

6883 6894 

6884 - `"program"`6895 - `"program"`

6885 6896 

6897 - `name: optional string`

6898 

6899 The name of the tool that produced the output.

6900 

6901 - `namespace: optional string`

6902 

6903 The namespace of the tool that produced the output.

6904 

6886 - `status: optional "in_progress" or "completed" or "incomplete"`6905 - `status: optional "in_progress" or "completed" or "incomplete"`

6887 6906 

6888 The status of the item. One of `in_progress`, `completed`, or6907 The status of the item. One of `in_progress`, `completed`, or

1055 1055 

1056 - `"incomplete"`1056 - `"incomplete"`

1057 1057 

1058 - `FunctionCallOutput object { call_id, output, type, 3 more }`1058 - `FunctionCallOutput object { call_id, output, type, 5 more }`

1059 1059 

1060 The output of a function tool call.1060 The output of a function tool call.

1061 1061 

1219 1219 

1220 - `"program"`1220 - `"program"`

1221 1221 

1222 - `name: optional string`

1223 

1224 The name of the tool that produced the output.

1225 

1226 - `namespace: optional string`

1227 

1228 The namespace of the tool that produced the output.

1229 

1222 - `status: optional "in_progress" or "completed" or "incomplete"`1230 - `status: optional "in_progress" or "completed" or "incomplete"`

1223 1231 

1224 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.1232 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

4571 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.4579 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

4572 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).4580 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).

4573 4581 

4574- `service_tier: optional "auto" or "default" or "flex" or 2 more`4582- `service_tier: optional "auto" or "default" or "flex" or 3 more`

4575 4583 

4576 Specifies the processing type used for serving the request.4584 Specifies the processing type used for serving the request.

4577 4585 

4578 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.4586 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

4579 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.4587 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

4580 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.4588 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

4589 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

4581 - When not set, the default behavior is 'auto'.4590 - When not set, the default behavior is 'auto'.

4582 4591 

4583 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.4592 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

4592 4601 

4593 - `"priority"`4602 - `"priority"`

4594 4603 

4604 - `"fast"`

4605 

4595- `store: optional boolean`4606- `store: optional boolean`

4596 4607 

4597 Whether to store the generated model response for later retrieval via4608 Whether to store the generated model response for later retrieval via

6805 6816 

6806 - `"incomplete"`6817 - `"incomplete"`

6807 6818 

6808 - `FunctionCallOutput object { call_id, output, type, 3 more }`6819 - `FunctionCallOutput object { call_id, output, type, 5 more }`

6809 6820 

6810 The output of a function tool call.6821 The output of a function tool call.

6811 6822 

6969 6980 

6970 - `"program"`6981 - `"program"`

6971 6982 

6983 - `name: optional string`

6984 

6985 The name of the tool that produced the output.

6986 

6987 - `namespace: optional string`

6988 

6989 The namespace of the tool that produced the output.

6990 

6972 - `status: optional "in_progress" or "completed" or "incomplete"`6991 - `status: optional "in_progress" or "completed" or "incomplete"`

6973 6992 

6974 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.6993 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

10242 10261 

10243 - `"incomplete"`10262 - `"incomplete"`

10244 10263 

10245 - `FunctionCallOutput object { id, call_id, output, 4 more }`10264 - `FunctionCallOutput object { id, call_id, output, 6 more }`

10246 10265 

10247 - `id: string`10266 - `id: string`

10248 10267 

10322 10341 

10323 The identifier of the actor that created the item.10342 The identifier of the actor that created the item.

10324 10343 

10344 - `name: optional string`

10345 

10346 The name of the tool that produced the output.

10347 

10348 - `namespace: optional string`

10349 

10350 The namespace of the tool that produced the output.

10351 

10325 - `WebSearchCall object { id, action, status, type }`10352 - `WebSearchCall object { id, action, status, type }`

10326 10353 

10327 The results of a web search tool call. See the10354 The results of a web search tool call. See the

14554 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.14581 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

14555 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).14582 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).

14556 14583 

14557 - `service_tier: optional "auto" or "default" or "flex" or 2 more`14584 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

14558 14585 

14559 Specifies the processing type used for serving the request.14586 Specifies the processing type used for serving the request.

14560 14587 

14561 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.14588 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

14562 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.14589 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

14563 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.14590 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

14591 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

14564 - When not set, the default behavior is 'auto'.14592 - When not set, the default behavior is 'auto'.

14565 14593 

14566 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.14594 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

14575 14603 

14576 - `"priority"`14604 - `"priority"`

14577 14605 

14606 - `"fast"`

14607 

14578 - `status: optional ResponseStatus`14608 - `status: optional ResponseStatus`

14579 14609 

14580 The status of the response generation. One of `completed`, `failed`,14610 The status of the response generation. One of `completed`, `failed`,

1107 1107 

1108 - `"incomplete"`1108 - `"incomplete"`

1109 1109 

1110 - `FunctionCallOutput object { call_id, output, type, 3 more }`1110 - `FunctionCallOutput object { call_id, output, type, 5 more }`

1111 1111 

1112 The output of a function tool call.1112 The output of a function tool call.

1113 1113 

1271 1271 

1272 - `"program"`1272 - `"program"`

1273 1273 

1274 - `name: optional string`

1275 

1276 The name of the tool that produced the output.

1277 

1278 - `namespace: optional string`

1279 

1280 The namespace of the tool that produced the output.

1281 

1274 - `status: optional "in_progress" or "completed" or "incomplete"`1282 - `status: optional "in_progress" or "completed" or "incomplete"`

1275 1283 

1276 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.1284 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

4544 4552 

4545 - `"incomplete"`4553 - `"incomplete"`

4546 4554 

4547 - `FunctionCallOutput object { id, call_id, output, 4 more }`4555 - `FunctionCallOutput object { id, call_id, output, 6 more }`

4548 4556 

4549 - `id: string`4557 - `id: string`

4550 4558 

4624 4632 

4625 The identifier of the actor that created the item.4633 The identifier of the actor that created the item.

4626 4634 

4635 - `name: optional string`

4636 

4637 The name of the tool that produced the output.

4638 

4639 - `namespace: optional string`

4640 

4641 The namespace of the tool that produced the output.

4642 

4627 - `WebSearchCall object { id, action, status, type }`4643 - `WebSearchCall object { id, action, status, type }`

4628 4644 

4629 The results of a web search tool call. See the4645 The results of a web search tool call. See the

8856 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.8872 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

8857 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).8873 The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](/docs/guides/safety-best-practices#safety-identifiers).

8858 8874 

8859 - `service_tier: optional "auto" or "default" or "flex" or 2 more`8875 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

8860 8876 

8861 Specifies the processing type used for serving the request.8877 Specifies the processing type used for serving the request.

8862 8878 

8863 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.8879 - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'.

8864 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.8880 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

8865 - If set to '[flex](/docs/guides/flex-processing)' or '[priority](https://openai.com/api-priority-processing/)', then the request will be processed with the corresponding service tier.8881 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

8882 - To opt-in to [Fast mode](/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request.

8866 - When not set, the default behavior is 'auto'.8883 - When not set, the default behavior is 'auto'.

8867 8884 

8868 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.8885 When the `service_tier` parameter is set, the response body will include the `service_tier` value based on the processing mode actually used to serve the request. This response value may be different from the value set in the parameter.

8877 8894 

8878 - `"priority"`8895 - `"priority"`

8879 8896 

8897 - `"fast"`

8898 

8880 - `status: optional ResponseStatus`8899 - `status: optional ResponseStatus`

8881 8900 

8882 The status of the response generation. One of `completed`, `failed`,8901 The status of the response generation. One of `completed`, `failed`,

976 976 

977 The namespace of the function to run.977 The namespace of the function to run.

978 978 

979 - `FunctionCallOutput object { id, call_id, output, 4 more }`979 - `FunctionCallOutput object { id, call_id, output, 6 more }`

980 980 

981 - `id: string`981 - `id: string`

982 982 

1056 1056 

1057 The identifier of the actor that created the item.1057 The identifier of the actor that created the item.

1058 1058 

1059 - `name: optional string`

1060 

1061 The name of the tool that produced the output.

1062 

1063 - `namespace: optional string`

1064 

1065 The namespace of the tool that produced the output.

1066 

1059 - `ToolSearchCall object { id, arguments, call_id, 4 more }`1067 - `ToolSearchCall object { id, arguments, call_id, 4 more }`

1060 1068 

1061 - `id: string`1069 - `id: string`

996 996 

997 - `"incomplete"`997 - `"incomplete"`

998 998 

999 - `FunctionCallOutput object { call_id, output, type, 3 more }`999 - `FunctionCallOutput object { call_id, output, type, 5 more }`

1000 1000 

1001 The output of a function tool call.1001 The output of a function tool call.

1002 1002 

1160 1160 

1161 - `"program"`1161 - `"program"`

1162 1162 

1163 - `name: optional string`

1164 

1165 The name of the tool that produced the output.

1166 

1167 - `namespace: optional string`

1168 

1169 The namespace of the tool that produced the output.

1170 

1163 - `status: optional "in_progress" or "completed" or "incomplete"`1171 - `status: optional "in_progress" or "completed" or "incomplete"`

1164 1172 

1165 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.1173 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.