Reference 2026-06-02 06:51 UTC to 2026-06-04 06:52 UTC

11 11 

12### Returns12### Returns

13 13 

14- `ChatCompletion object { id, choices, created, 5 more }`14- `ChatCompletion object { id, choices, created, 6 more }`

15 15 

16 Represents a chat completion response returned by model, based on the provided input.16 Represents a chat completion response returned by model, based on the provided input.

17 17 

29 `length` if the maximum number of tokens specified in the request was reached,29 `length` if the maximum number of tokens specified in the request was reached,

30 `content_filter` if content was omitted due to a flag from our content filters,30 `content_filter` if content was omitted due to a flag from our content filters,

31 `tool_calls` if the model called a tool, or `function_call` (deprecated) if the model called a function.31 `tool_calls` if the model called a tool, or `function_call` (deprecated) if the model called a function.

32 Read the [Model Spec](https://model-spec.openai.com/2025-12-18.html) for more.

32 33 

33 - `"stop"`34 - `"stop"`

34 35 

255 256 

256 - `"chat.completion"`257 - `"chat.completion"`

257 258 

259 - `moderation: optional object { input, output }`

260 

261 Moderation results for the request input and generated output, if moderated

262 completions were requested.

263 

264 - `input: object { model, results, type } or object { code, message, type }`

265 

266 Moderation for the request input.

267 

268 - `ModerationResults object { model, results, type }`

269 

270 Successful moderation results for the request input or generated output.

271 

272 - `model: string`

273 

274 The moderation model used to generate the results.

275 

276 - `results: array of object { categories, category_applied_input_types, category_scores, 3 more }`

277 

278 A list of moderation results.

279 

280 - `categories: map[boolean]`

281 

282 A dictionary of moderation categories to booleans, True if the input is flagged under this category.

283 

284 - `category_applied_input_types: map[array of "text" or "image"]`

285 

286 Which modalities of input are reflected by the score for each category.

287 

288 - `"text"`

289 

290 - `"image"`

291 

292 - `category_scores: map[number]`

293 

294 A dictionary of moderation categories to scores.

295 

296 - `flagged: boolean`

297 

298 A boolean indicating whether the content was flagged by any category.

299 

300 - `model: string`

301 

302 The moderation model that produced this result.

303 

304 - `type: "moderation_result"`

305 

306 The object type, which was always `moderation_result` for successful moderation results.

307 

308 - `"moderation_result"`

309 

310 - `type: "moderation_results"`

311 

312 The object type, which is always `moderation_results`.

313 

314 - `"moderation_results"`

315 

316 - `Error object { code, message, type }`

317 

318 An error produced while attempting moderation.

319 

320 - `code: string`

321 

322 The error code.

323 

324 - `message: string`

325 

326 The error message.

327 

328 - `type: "error"`

329 

330 The object type, which is always `error`.

331 

332 - `"error"`

333 

334 - `output: object { model, results, type } or object { code, message, type }`

335 

336 Moderation for the generated output.

337 

338 - `ModerationResults object { model, results, type }`

339 

340 Successful moderation results for the request input or generated output.

341 

342 - `model: string`

343 

344 The moderation model used to generate the results.

345 

346 - `results: array of object { categories, category_applied_input_types, category_scores, 3 more }`

347 

348 A list of moderation results.

349 

350 - `categories: map[boolean]`

351 

352 A dictionary of moderation categories to booleans, True if the input is flagged under this category.

353 

354 - `category_applied_input_types: map[array of "text" or "image"]`

355 

356 Which modalities of input are reflected by the score for each category.

357 

358 - `"text"`

359 

360 - `"image"`

361 

362 - `category_scores: map[number]`

363 

364 A dictionary of moderation categories to scores.

365 

366 - `flagged: boolean`

367 

368 A boolean indicating whether the content was flagged by any category.

369 

370 - `model: string`

371 

372 The moderation model that produced this result.

373 

374 - `type: "moderation_result"`

375 

376 The object type, which was always `moderation_result` for successful moderation results.

377 

378 - `"moderation_result"`

379 

380 - `type: "moderation_results"`

381 

382 The object type, which is always `moderation_results`.

383 

384 - `"moderation_results"`

385 

386 - `Error object { code, message, type }`

387 

388 An error produced while attempting moderation.

389 

390 - `code: string`

391 

392 The error code.

393 

394 - `message: string`

395 

396 The error message.

397 

398 - `type: "error"`

399 

400 The object type, which is always `error`.

401 

402 - `"error"`

403 

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

259 405 

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

430 "created": 0,576 "created": 0,

431 "model": "model",577 "model": "model",

432 "object": "chat.completion",578 "object": "chat.completion",

579 "moderation": {

580 "input": {

581 "model": "model",

582 "results": [

583 {

584 "categories": {

585 "foo": true

586 },

587 "category_applied_input_types": {

588 "foo": [

589 "text"

590 ]

591 },

592 "category_scores": {

593 "foo": 0

594 },

595 "flagged": true,

596 "model": "model",

597 "type": "moderation_result"

598 }

599 ],

600 "type": "moderation_results"

601 },

602 "output": {

603 "model": "model",

604 "results": [

605 {

606 "categories": {

607 "foo": true

608 },

609 "category_applied_input_types": {

610 "foo": [

611 "text"

612 ]

613 },

614 "category_scores": {

615 "foo": 0

616 },

617 "flagged": true,

618 "model": "model",

619 "type": "moderation_result"

620 }

621 ],

622 "type": "moderation_results"

623 }

624 },

433 "service_tier": "auto",625 "service_tier": "auto",

434 "system_fingerprint": "system_fingerprint",626 "system_fingerprint": "system_fingerprint",

435 "usage": {627 "usage": {

12 12 

13### Returns13### Returns

14 14 

15- `Response object { id, created_at, error, 30 more }`15- `Response object { id, created_at, error, 31 more }`

16 16 

17 - `id: string`17 - `id: string`

18 18 

7537 7537 

7538 The maximum number of total calls to built-in tools that can be processed in a response. This maximum number applies across all built-in tool calls, not per individual tool. Any further attempts to call a tool by the model will be ignored.7538 The maximum number of total calls to built-in tools that can be processed in a response. This maximum number applies across all built-in tool calls, not per individual tool. Any further attempts to call a tool by the model will be ignored.

7539 7539 

7540 - `moderation: optional object { input, output }`

7541 

7542 Moderation results for the response input and output, if moderated completions were requested.

7543 

7544 - `input: object { categories, category_applied_input_types, category_scores, 3 more } or object { code, message, type }`

7545 

7546 Moderation for the response input.

7547 

7548 - `ModerationResult object { categories, category_applied_input_types, category_scores, 3 more }`

7549 

7550 A moderation result produced for the response input or output.

7551 

7552 - `categories: map[boolean]`

7553 

7554 A dictionary of moderation categories to booleans, True if the input is flagged under this category.

7555 

7556 - `category_applied_input_types: map[array of "text" or "image"]`

7557 

7558 Which modalities of input are reflected by the score for each category.

7559 

7560 - `"text"`

7561 

7562 - `"image"`

7563 

7564 - `category_scores: map[number]`

7565 

7566 A dictionary of moderation categories to scores.

7567 

7568 - `flagged: boolean`

7569 

7570 A boolean indicating whether the content was flagged by any category.

7571 

7572 - `model: string`

7573 

7574 The moderation model that produced this result.

7575 

7576 - `type: "moderation_result"`

7577 

7578 The object type, which was always `moderation_result` for successful moderation results.

7579 

7580 - `"moderation_result"`

7581 

7582 - `Error object { code, message, type }`

7583 

7584 An error produced while attempting moderation for the response input or output.

7585 

7586 - `code: string`

7587 

7588 The error code.

7589 

7590 - `message: string`

7591 

7592 The error message.

7593 

7594 - `type: "error"`

7595 

7596 The object type, which was always `error` for moderation failures.

7597 

7598 - `"error"`

7599 

7600 - `output: object { categories, category_applied_input_types, category_scores, 3 more } or object { code, message, type }`

7601 

7602 Moderation for the response output.

7603 

7604 - `ModerationResult object { categories, category_applied_input_types, category_scores, 3 more }`

7605 

7606 A moderation result produced for the response input or output.

7607 

7608 - `categories: map[boolean]`

7609 

7610 A dictionary of moderation categories to booleans, True if the input is flagged under this category.

7611 

7612 - `category_applied_input_types: map[array of "text" or "image"]`

7613 

7614 Which modalities of input are reflected by the score for each category.

7615 

7616 - `"text"`

7617 

7618 - `"image"`

7619 

7620 - `category_scores: map[number]`

7621 

7622 A dictionary of moderation categories to scores.

7623 

7624 - `flagged: boolean`

7625 

7626 A boolean indicating whether the content was flagged by any category.

7627 

7628 - `model: string`

7629 

7630 The moderation model that produced this result.

7631 

7632 - `type: "moderation_result"`

7633 

7634 The object type, which was always `moderation_result` for successful moderation results.

7635 

7636 - `"moderation_result"`

7637 

7638 - `Error object { code, message, type }`

7639 

7640 An error produced while attempting moderation for the response input or output.

7641 

7642 - `code: string`

7643 

7644 The error code.

7645 

7646 - `message: string`

7647 

7648 The error message.

7649 

7650 - `type: "error"`

7651 

7652 The object type, which was always `error` for moderation failures.

7653 

7654 - `"error"`

7655 

7540 - `output_text: optional string`7656 - `output_text: optional string`

7541 7657 

7542 SDK-only convenience property that contains the aggregated text output7658 SDK-only convenience property that contains the aggregated text output

7948 },8064 },

7949 "max_output_tokens": 0,8065 "max_output_tokens": 0,

7950 "max_tool_calls": 0,8066 "max_tool_calls": 0,

8067 "moderation": {

8068 "input": {

8069 "categories": {

8070 "foo": true

8071 },

8072 "category_applied_input_types": {

8073 "foo": [

8074 "text"

8075 ]

8076 },

8077 "category_scores": {

8078 "foo": 0

8079 },

8080 "flagged": true,

8081 "model": "model",

8082 "type": "moderation_result"

8083 },

8084 "output": {

8085 "categories": {

8086 "foo": true

8087 },

8088 "category_applied_input_types": {

8089 "foo": [

8090 "text"

8091 ]

8092 },

8093 "category_scores": {

8094 "foo": 0

8095 },

8096 "flagged": true,

8097 "model": "model",

8098 "type": "moderation_result"

8099 }

8100 },

7951 "output_text": "output_text",8101 "output_text": "output_text",

7952 "previous_response_id": "previous_response_id",8102 "previous_response_id": "previous_response_id",

7953 "prompt": {8103 "prompt": {

3876 3876 

3877 - `"gpt-5.1-codex-max"`3877 - `"gpt-5.1-codex-max"`

3878 3878 

3879- `moderation: optional object { model }`

3880 

3881 Configuration for running moderation on the input and output of this response.

3882 

3883 - `model: string`

3884 

3885 The moderation model to use for moderated completions, e.g. 'omni-moderation-latest'.

3886 

3879- `parallel_tool_calls: optional boolean`3887- `parallel_tool_calls: optional boolean`

3880 3888 

3881 Whether to allow the model to run tool calls in parallel.3889 Whether to allow the model to run tool calls in parallel.

5089 5097 

5090### Returns5098### Returns

5091 5099 

5092- `Response object { id, created_at, error, 30 more }`5100- `Response object { id, created_at, error, 31 more }`

5093 5101 

5094 - `id: string`5102 - `id: string`

5095 5103 

12614 12622 

12615 The maximum number of total calls to built-in tools that can be processed in a response. This maximum number applies across all built-in tool calls, not per individual tool. Any further attempts to call a tool by the model will be ignored.12623 The maximum number of total calls to built-in tools that can be processed in a response. This maximum number applies across all built-in tool calls, not per individual tool. Any further attempts to call a tool by the model will be ignored.

12616 12624 

12625 - `moderation: optional object { input, output }`

12626 

12627 Moderation results for the response input and output, if moderated completions were requested.

12628 

12629 - `input: object { categories, category_applied_input_types, category_scores, 3 more } or object { code, message, type }`

12630 

12631 Moderation for the response input.

12632 

12633 - `ModerationResult object { categories, category_applied_input_types, category_scores, 3 more }`

12634 

12635 A moderation result produced for the response input or output.

12636 

12637 - `categories: map[boolean]`

12638 

12639 A dictionary of moderation categories to booleans, True if the input is flagged under this category.

12640 

12641 - `category_applied_input_types: map[array of "text" or "image"]`

12642 

12643 Which modalities of input are reflected by the score for each category.

12644 

12645 - `"text"`

12646 

12647 - `"image"`

12648 

12649 - `category_scores: map[number]`

12650 

12651 A dictionary of moderation categories to scores.

12652 

12653 - `flagged: boolean`

12654 

12655 A boolean indicating whether the content was flagged by any category.

12656 

12657 - `model: string`

12658 

12659 The moderation model that produced this result.

12660 

12661 - `type: "moderation_result"`

12662 

12663 The object type, which was always `moderation_result` for successful moderation results.

12664 

12665 - `"moderation_result"`

12666 

12667 - `Error object { code, message, type }`

12668 

12669 An error produced while attempting moderation for the response input or output.

12670 

12671 - `code: string`

12672 

12673 The error code.

12674 

12675 - `message: string`

12676 

12677 The error message.

12678 

12679 - `type: "error"`

12680 

12681 The object type, which was always `error` for moderation failures.

12682 

12683 - `"error"`

12684 

12685 - `output: object { categories, category_applied_input_types, category_scores, 3 more } or object { code, message, type }`

12686 

12687 Moderation for the response output.

12688 

12689 - `ModerationResult object { categories, category_applied_input_types, category_scores, 3 more }`

12690 

12691 A moderation result produced for the response input or output.

12692 

12693 - `categories: map[boolean]`

12694 

12695 A dictionary of moderation categories to booleans, True if the input is flagged under this category.

12696 

12697 - `category_applied_input_types: map[array of "text" or "image"]`

12698 

12699 Which modalities of input are reflected by the score for each category.

12700 

12701 - `"text"`

12702 

12703 - `"image"`

12704 

12705 - `category_scores: map[number]`

12706 

12707 A dictionary of moderation categories to scores.

12708 

12709 - `flagged: boolean`

12710 

12711 A boolean indicating whether the content was flagged by any category.

12712 

12713 - `model: string`

12714 

12715 The moderation model that produced this result.

12716 

12717 - `type: "moderation_result"`

12718 

12719 The object type, which was always `moderation_result` for successful moderation results.

12720 

12721 - `"moderation_result"`

12722 

12723 - `Error object { code, message, type }`

12724 

12725 An error produced while attempting moderation for the response input or output.

12726 

12727 - `code: string`

12728 

12729 The error code.

12730 

12731 - `message: string`

12732 

12733 The error message.

12734 

12735 - `type: "error"`

12736 

12737 The object type, which was always `error` for moderation failures.

12738 

12739 - `"error"`

12740 

12617 - `output_text: optional string`12741 - `output_text: optional string`

12618 12742 

12619 SDK-only convenience property that contains the aggregated text output12743 SDK-only convenience property that contains the aggregated text output

13033 },13157 },

13034 "max_output_tokens": 0,13158 "max_output_tokens": 0,

13035 "max_tool_calls": 0,13159 "max_tool_calls": 0,

13160 "moderation": {

13161 "input": {

13162 "categories": {

13163 "foo": true

13164 },

13165 "category_applied_input_types": {

13166 "foo": [

13167 "text"

13168 ]

13169 },

13170 "category_scores": {

13171 "foo": 0

13172 },

13173 "flagged": true,

13174 "model": "model",

13175 "type": "moderation_result"

13176 },

13177 "output": {

13178 "categories": {

13179 "foo": true

13180 },

13181 "category_applied_input_types": {

13182 "foo": [

13183 "text"

13184 ]

13185 },

13186 "category_scores": {

13187 "foo": 0

13188 },

13189 "flagged": true,

13190 "model": "model",

13191 "type": "moderation_result"

13192 }

13193 },

13036 "output_text": "output_text",13194 "output_text": "output_text",

13037 "previous_response_id": "previous_response_id",13195 "previous_response_id": "previous_response_id",

13038 "prompt": {13196 "prompt": {

56 56 

57### Returns57### Returns

58 58 

59- `Response object { id, created_at, error, 30 more }`59- `Response object { id, created_at, error, 31 more }`

60 60 

61 - `id: string`61 - `id: string`

62 62 

7581 7581 

7582 The maximum number of total calls to built-in tools that can be processed in a response. This maximum number applies across all built-in tool calls, not per individual tool. Any further attempts to call a tool by the model will be ignored.7582 The maximum number of total calls to built-in tools that can be processed in a response. This maximum number applies across all built-in tool calls, not per individual tool. Any further attempts to call a tool by the model will be ignored.

7583 7583 

7584 - `moderation: optional object { input, output }`

7585 

7586 Moderation results for the response input and output, if moderated completions were requested.

7587 

7588 - `input: object { categories, category_applied_input_types, category_scores, 3 more } or object { code, message, type }`

7589 

7590 Moderation for the response input.

7591 

7592 - `ModerationResult object { categories, category_applied_input_types, category_scores, 3 more }`

7593 

7594 A moderation result produced for the response input or output.

7595 

7596 - `categories: map[boolean]`

7597 

7598 A dictionary of moderation categories to booleans, True if the input is flagged under this category.

7599 

7600 - `category_applied_input_types: map[array of "text" or "image"]`

7601 

7602 Which modalities of input are reflected by the score for each category.

7603 

7604 - `"text"`

7605 

7606 - `"image"`

7607 

7608 - `category_scores: map[number]`

7609 

7610 A dictionary of moderation categories to scores.

7611 

7612 - `flagged: boolean`

7613 

7614 A boolean indicating whether the content was flagged by any category.

7615 

7616 - `model: string`

7617 

7618 The moderation model that produced this result.

7619 

7620 - `type: "moderation_result"`

7621 

7622 The object type, which was always `moderation_result` for successful moderation results.

7623 

7624 - `"moderation_result"`

7625 

7626 - `Error object { code, message, type }`

7627 

7628 An error produced while attempting moderation for the response input or output.

7629 

7630 - `code: string`

7631 

7632 The error code.

7633 

7634 - `message: string`

7635 

7636 The error message.

7637 

7638 - `type: "error"`

7639 

7640 The object type, which was always `error` for moderation failures.

7641 

7642 - `"error"`

7643 

7644 - `output: object { categories, category_applied_input_types, category_scores, 3 more } or object { code, message, type }`

7645 

7646 Moderation for the response output.

7647 

7648 - `ModerationResult object { categories, category_applied_input_types, category_scores, 3 more }`

7649 

7650 A moderation result produced for the response input or output.

7651 

7652 - `categories: map[boolean]`

7653 

7654 A dictionary of moderation categories to booleans, True if the input is flagged under this category.

7655 

7656 - `category_applied_input_types: map[array of "text" or "image"]`

7657 

7658 Which modalities of input are reflected by the score for each category.

7659 

7660 - `"text"`

7661 

7662 - `"image"`

7663 

7664 - `category_scores: map[number]`

7665 

7666 A dictionary of moderation categories to scores.

7667 

7668 - `flagged: boolean`

7669 

7670 A boolean indicating whether the content was flagged by any category.

7671 

7672 - `model: string`

7673 

7674 The moderation model that produced this result.

7675 

7676 - `type: "moderation_result"`

7677 

7678 The object type, which was always `moderation_result` for successful moderation results.

7679 

7680 - `"moderation_result"`

7681 

7682 - `Error object { code, message, type }`

7683 

7684 An error produced while attempting moderation for the response input or output.

7685 

7686 - `code: string`

7687 

7688 The error code.

7689 

7690 - `message: string`

7691 

7692 The error message.

7693 

7694 - `type: "error"`

7695 

7696 The object type, which was always `error` for moderation failures.

7697 

7698 - `"error"`

7699 

7584 - `output_text: optional string`7700 - `output_text: optional string`

7585 7701 

7586 SDK-only convenience property that contains the aggregated text output7702 SDK-only convenience property that contains the aggregated text output

7991 },8107 },

7992 "max_output_tokens": 0,8108 "max_output_tokens": 0,

7993 "max_tool_calls": 0,8109 "max_tool_calls": 0,

8110 "moderation": {

8111 "input": {

8112 "categories": {

8113 "foo": true

8114 },

8115 "category_applied_input_types": {

8116 "foo": [

8117 "text"

8118 ]

8119 },

8120 "category_scores": {

8121 "foo": 0

8122 },

8123 "flagged": true,

8124 "model": "model",

8125 "type": "moderation_result"

8126 },

8127 "output": {

8128 "categories": {

8129 "foo": true

8130 },

8131 "category_applied_input_types": {

8132 "foo": [

8133 "text"

8134 ]

8135 },

8136 "category_scores": {

8137 "foo": 0

8138 },

8139 "flagged": true,

8140 "model": "model",

8141 "type": "moderation_result"

8142 }

8143 },

7994 "output_text": "output_text",8144 "output_text": "output_text",

7995 "previous_response_id": "previous_response_id",8145 "previous_response_id": "previous_response_id",

7996 "prompt": {8146 "prompt": {