Reference 2026-07-22 20:02 UTC to 2026-07-23 18:00 UTC

591 - `context: optional "auto" or "current_turn" or "all_turns"`591 - `context: optional "auto" or "current_turn" or "all_turns"`

592 592 

593 Controls which reasoning items are rendered back to the model on later turns.593 Controls which reasoning items are rendered back to the model on later turns.

594 If omitted or set to `auto`, the model determines the context mode. The

595 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

596 `current_turn`.

597 

594 When returned on a response, this is the effective reasoning context mode598 When returned on a response, this is the effective reasoning context mode

595 used for the response.599 used for the response.

596 600 

30 30 

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

32 32 

33 The identifier of the API key.

34 

33 - `created_at: number`35 - `created_at: number`

34 36 

37 The Unix timestamp (in seconds) when the API key was created.

38 

35 - `name: string`39 - `name: string`

36 40 

41 The name of the API key.

42 

37 - `object: "organization.project.service_account.api_key"`43 - `object: "organization.project.service_account.api_key"`

38 44 

39 The object type, which is always `organization.project.service_account.api_key`45 The object type, which is always `organization.project.service_account.api_key`

40 46 

41 - `value: string`47 - `value: string`

42 48 

49 The unredacted API key value.

50 

43### Example51### Example

44 52 

45```cli53```cli

1## Delete project spend limit

2 

3`$ openai admin:organization:projects:spend-limit delete`

4 

5**delete** `/organization/projects/{project_id}/spend_limit`

6 

7Delete a project's hard spend limit.

8 

9### Parameters

10 

11- `--project-id: string`

12 

13 The ID of the project whose hard spend limit is being managed.

14 

15### Returns

16 

17- `project_spend_limit_deleted: object { deleted, object }`

18 

19 Confirmation payload returned after deleting a project hard spend limit.

20 

21 - `deleted: boolean`

22 

23 Whether the hard spend limit was deleted.

24 

25 - `object: "project.spend_limit.deleted"`

26 

27 The object type, which is always `project.spend_limit.deleted`.

28 

29### Example

30 

31```cli

32openai admin:organization:projects:spend-limit delete \

33 --admin-api-key 'My Admin API Key' \

34 --project-id proj_123

35```

36 

37#### Response

38 

39```json

40{

41 "deleted": true,

42 "object": "project.spend_limit.deleted"

43}

44```

1## Retrieve project spend limit

2 

3`$ openai admin:organization:projects:spend-limit retrieve`

4 

5**get** `/organization/projects/{project_id}/spend_limit`

6 

7Get a project's hard spend limit.

8 

9### Parameters

10 

11- `--project-id: string`

12 

13 The ID of the project whose hard spend limit is being managed.

14 

15### Returns

16 

17- `project_spend_limit: object { currency, enforcement, interval, 2 more }`

18 

19 Represents a hard spend limit configured at the project level.

20 

21 - `currency: string or "USD"`

22 

23 The currency for the threshold amount. Currently, only `USD` is supported.

24 

25 - `"USD"`

26 

27 - `enforcement: object { status }`

28 

29 The current enforcement state of the hard spend limit.

30 

31 - `status: string or "inactive" or "enforcing"`

32 

33 Whether the hard spend limit is currently enforcing.

34 

35 - `"inactive"`

36 

37 - `"enforcing"`

38 

39 - `interval: string or "month"`

40 

41 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

42 

43 - `"month"`

44 

45 - `object: "project.spend_limit"`

46 

47 The object type, which is always `project.spend_limit`.

48 

49 - `threshold_amount: number`

50 

51 The hard spend limit amount, in cents.

52 

53### Example

54 

55```cli

56openai admin:organization:projects:spend-limit retrieve \

57 --admin-api-key 'My Admin API Key' \

58 --project-id proj_123

59```

60 

61#### Response

62 

63```json

64{

65 "currency": "USD",

66 "enforcement": {

67 "status": "inactive"

68 },

69 "interval": "month",

70 "object": "project.spend_limit",

71 "threshold_amount": 0

72}

73```

1## Update project spend limit

2 

3`$ openai admin:organization:projects:spend-limit update`

4 

5**post** `/organization/projects/{project_id}/spend_limit`

6 

7Create or replace a project's hard spend limit.

8 

9### Parameters

10 

11- `--project-id: string`

12 

13 The ID of the project whose hard spend limit is being managed.

14 

15- `--currency: "USD"`

16 

17 The currency for the threshold amount. Currently, only `USD` is supported.

18 

19- `--interval: "month"`

20 

21 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

22 

23- `--threshold-amount: number`

24 

25 The hard spend limit amount, in cents.

26 

27### Returns

28 

29- `project_spend_limit: object { currency, enforcement, interval, 2 more }`

30 

31 Represents a hard spend limit configured at the project level.

32 

33 - `currency: string or "USD"`

34 

35 The currency for the threshold amount. Currently, only `USD` is supported.

36 

37 - `"USD"`

38 

39 - `enforcement: object { status }`

40 

41 The current enforcement state of the hard spend limit.

42 

43 - `status: string or "inactive" or "enforcing"`

44 

45 Whether the hard spend limit is currently enforcing.

46 

47 - `"inactive"`

48 

49 - `"enforcing"`

50 

51 - `interval: string or "month"`

52 

53 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

54 

55 - `"month"`

56 

57 - `object: "project.spend_limit"`

58 

59 The object type, which is always `project.spend_limit`.

60 

61 - `threshold_amount: number`

62 

63 The hard spend limit amount, in cents.

64 

65### Example

66 

67```cli

68openai admin:organization:projects:spend-limit update \

69 --admin-api-key 'My Admin API Key' \

70 --project-id proj_123 \

71 --currency USD \

72 --interval month \

73 --threshold-amount 1

74```

75 

76#### Response

77 

78```json

79{

80 "currency": "USD",

81 "enforcement": {

82 "status": "inactive"

83 },

84 "interval": "month",

85 "object": "project.spend_limit",

86 "threshold_amount": 0

87}

88```

1## Delete organization spend limit

2 

3`$ openai admin:organization:spend-limit delete`

4 

5**delete** `/organization/spend_limit`

6 

7Delete the organization's hard spend limit.

8 

9### Returns

10 

11- `organization_spend_limit_deleted: object { deleted, object }`

12 

13 Confirmation payload returned after deleting an organization hard spend limit.

14 

15 - `deleted: boolean`

16 

17 Whether the hard spend limit was deleted.

18 

19 - `object: "organization.spend_limit.deleted"`

20 

21 The object type, which is always `organization.spend_limit.deleted`.

22 

23### Example

24 

25```cli

26openai admin:organization:spend-limit delete \

27 --admin-api-key 'My Admin API Key'

28```

29 

30#### Response

31 

32```json

33{

34 "deleted": true,

35 "object": "organization.spend_limit.deleted"

36}

37```

1## Retrieve organization spend limit

2 

3`$ openai admin:organization:spend-limit retrieve`

4 

5**get** `/organization/spend_limit`

6 

7Get the organization's hard spend limit.

8 

9### Returns

10 

11- `organization_spend_limit: object { currency, enforcement, interval, 2 more }`

12 

13 Represents a hard spend limit configured at the organization level.

14 

15 - `currency: string or "USD"`

16 

17 The currency for the threshold amount. Currently, only `USD` is supported.

18 

19 - `"USD"`

20 

21 - `enforcement: object { status }`

22 

23 The current enforcement state of the hard spend limit.

24 

25 - `status: string or "inactive" or "enforcing"`

26 

27 Whether the hard spend limit is currently enforcing.

28 

29 - `"inactive"`

30 

31 - `"enforcing"`

32 

33 - `interval: string or "month"`

34 

35 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

36 

37 - `"month"`

38 

39 - `object: "organization.spend_limit"`

40 

41 The object type, which is always `organization.spend_limit`.

42 

43 - `threshold_amount: number`

44 

45 The hard spend limit amount, in cents.

46 

47### Example

48 

49```cli

50openai admin:organization:spend-limit retrieve \

51 --admin-api-key 'My Admin API Key'

52```

53 

54#### Response

55 

56```json

57{

58 "currency": "USD",

59 "enforcement": {

60 "status": "inactive"

61 },

62 "interval": "month",

63 "object": "organization.spend_limit",

64 "threshold_amount": 0

65}

66```

1## Update organization spend limit

2 

3`$ openai admin:organization:spend-limit update`

4 

5**post** `/organization/spend_limit`

6 

7Create or replace the organization's hard spend limit.

8 

9### Parameters

10 

11- `--currency: "USD"`

12 

13 The currency for the threshold amount. Currently, only `USD` is supported.

14 

15- `--interval: "month"`

16 

17 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

18 

19- `--threshold-amount: number`

20 

21 The hard spend limit amount, in cents.

22 

23### Returns

24 

25- `organization_spend_limit: object { currency, enforcement, interval, 2 more }`

26 

27 Represents a hard spend limit configured at the organization level.

28 

29 - `currency: string or "USD"`

30 

31 The currency for the threshold amount. Currently, only `USD` is supported.

32 

33 - `"USD"`

34 

35 - `enforcement: object { status }`

36 

37 The current enforcement state of the hard spend limit.

38 

39 - `status: string or "inactive" or "enforcing"`

40 

41 Whether the hard spend limit is currently enforcing.

42 

43 - `"inactive"`

44 

45 - `"enforcing"`

46 

47 - `interval: string or "month"`

48 

49 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

50 

51 - `"month"`

52 

53 - `object: "organization.spend_limit"`

54 

55 The object type, which is always `organization.spend_limit`.

56 

57 - `threshold_amount: number`

58 

59 The hard spend limit amount, in cents.

60 

61### Example

62 

63```cli

64openai admin:organization:spend-limit update \

65 --admin-api-key 'My Admin API Key' \

66 --currency USD \

67 --interval month \

68 --threshold-amount 1

69```

70 

71#### Response

72 

73```json

74{

75 "currency": "USD",

76 "enforcement": {

77 "status": "inactive"

78 },

79 "interval": "month",

80 "object": "organization.spend_limit",

81 "threshold_amount": 0

82}

83```

34 34 

35 An error object returned when the model fails to generate a Response.35 An error object returned when the model fails to generate a Response.

36 36 

37 - `code: "server_error" or "rate_limit_exceeded" or "invalid_prompt" or 16 more`37 - `code: "server_error" or "rate_limit_exceeded" or "invalid_prompt" or 17 more`

38 38 

39 The error code for the response.39 The error code for the response.

40 40 

44 44 

45 - `"invalid_prompt"`45 - `"invalid_prompt"`

46 46 

47 - `"data_residency_mismatch"`

48 

47 - `"bio_policy"`49 - `"bio_policy"`

48 50 

49 - `"vector_store_timeout"`51 - `"vector_store_timeout"`

6101 - `context: optional "auto" or "current_turn" or "all_turns"`6103 - `context: optional "auto" or "current_turn" or "all_turns"`

6102 6104 

6103 Controls which reasoning items are rendered back to the model on later turns.6105 Controls which reasoning items are rendered back to the model on later turns.

6106 If omitted or set to `auto`, the model determines the context mode. The

6107 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

6108 `current_turn`.

6109 

6104 When returned on a response, this is the effective reasoning context mode6110 When returned on a response, this is the effective reasoning context mode

6105 used for the response.6111 used for the response.

6106 6112 

6286 6292 

6287 Constrains the verbosity of the model's response. Lower values will result in6293 Constrains the verbosity of the model's response. Lower values will result in

6288 more concise responses, while higher values will result in more verbose responses.6294 more concise responses, while higher values will result in more verbose responses.

6289 Currently supported values are `low`, `medium`, and `high`.6295 Currently supported values are `low`, `medium`, and `high`. The default is

6296 `medium`.

6290 6297 

6291 - `"low"`6298 - `"low"`

6292 6299 

253 253 

254 An error object returned when the model fails to generate a Response.254 An error object returned when the model fails to generate a Response.

255 255 

256 - `code: "server_error" or "rate_limit_exceeded" or "invalid_prompt" or 16 more`256 - `code: "server_error" or "rate_limit_exceeded" or "invalid_prompt" or 17 more`

257 257 

258 The error code for the response.258 The error code for the response.

259 259 

263 263 

264 - `"invalid_prompt"`264 - `"invalid_prompt"`

265 265 

266 - `"data_residency_mismatch"`

267 

266 - `"bio_policy"`268 - `"bio_policy"`

267 269 

268 - `"vector_store_timeout"`270 - `"vector_store_timeout"`

6320 - `context: optional "auto" or "current_turn" or "all_turns"`6322 - `context: optional "auto" or "current_turn" or "all_turns"`

6321 6323 

6322 Controls which reasoning items are rendered back to the model on later turns.6324 Controls which reasoning items are rendered back to the model on later turns.

6325 If omitted or set to `auto`, the model determines the context mode. The

6326 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

6327 `current_turn`.

6328 

6323 When returned on a response, this is the effective reasoning context mode6329 When returned on a response, this is the effective reasoning context mode

6324 used for the response.6330 used for the response.

6325 6331 

6505 6511 

6506 Constrains the verbosity of the model's response. Lower values will result in6512 Constrains the verbosity of the model's response. Lower values will result in

6507 more concise responses, while higher values will result in more verbose responses.6513 more concise responses, while higher values will result in more verbose responses.

6508 Currently supported values are `low`, `medium`, and `high`.6514 Currently supported values are `low`, `medium`, and `high`. The default is

6515 `medium`.

6509 6516 

6510 - `"low"`6517 - `"low"`

6511 6518 

51 51 

52 An error object returned when the model fails to generate a Response.52 An error object returned when the model fails to generate a Response.

53 53 

54 - `code: "server_error" or "rate_limit_exceeded" or "invalid_prompt" or 16 more`54 - `code: "server_error" or "rate_limit_exceeded" or "invalid_prompt" or 17 more`

55 55 

56 The error code for the response.56 The error code for the response.

57 57 

61 61 

62 - `"invalid_prompt"`62 - `"invalid_prompt"`

63 63 

64 - `"data_residency_mismatch"`

65 

64 - `"bio_policy"`66 - `"bio_policy"`

65 67 

66 - `"vector_store_timeout"`68 - `"vector_store_timeout"`

6118 - `context: optional "auto" or "current_turn" or "all_turns"`6120 - `context: optional "auto" or "current_turn" or "all_turns"`

6119 6121 

6120 Controls which reasoning items are rendered back to the model on later turns.6122 Controls which reasoning items are rendered back to the model on later turns.

6123 If omitted or set to `auto`, the model determines the context mode. The

6124 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

6125 `current_turn`.

6126 

6121 When returned on a response, this is the effective reasoning context mode6127 When returned on a response, this is the effective reasoning context mode

6122 used for the response.6128 used for the response.

6123 6129 

6303 6309 

6304 Constrains the verbosity of the model's response. Lower values will result in6310 Constrains the verbosity of the model's response. Lower values will result in

6305 more concise responses, while higher values will result in more verbose responses.6311 more concise responses, while higher values will result in more verbose responses.

6306 Currently supported values are `low`, `medium`, and `high`.6312 Currently supported values are `low`, `medium`, and `high`. The default is

6313 `medium`.

6307 6314 

6308 - `"low"`6315 - `"low"`

6309 6316 

284 284 

285 Constrains the verbosity of the model's response. Lower values will result in285 Constrains the verbosity of the model's response. Lower values will result in

286 more concise responses, while higher values will result in more verbose responses.286 more concise responses, while higher values will result in more verbose responses.

287 Currently supported values are `low`, `medium`, and `high`.287 Currently supported values are `low`, `medium`, and `high`. The default is

288 `medium`.

288 289 

289- `--web-search-options: optional object { search_context_size, user_location }`290- `--web-search-options: optional object { search_context_size, user_location }`

290 291 

30 30 

31 An error object returned when the model fails to generate a Response.31 An error object returned when the model fails to generate a Response.

32 32 

33 - `code: "server_error" or "rate_limit_exceeded" or "invalid_prompt" or 16 more`33 - `code: "server_error" or "rate_limit_exceeded" or "invalid_prompt" or 17 more`

34 34 

35 The error code for the response.35 The error code for the response.

36 36 

40 40 

41 - `"invalid_prompt"`41 - `"invalid_prompt"`

42 42 

43 - `"data_residency_mismatch"`

44 

43 - `"bio_policy"`45 - `"bio_policy"`

44 46 

45 - `"vector_store_timeout"`47 - `"vector_store_timeout"`

5273 - `context: optional "auto" or "current_turn" or "all_turns"`5275 - `context: optional "auto" or "current_turn" or "all_turns"`

5274 5276 

5275 Controls which reasoning items are rendered back to the model on later turns.5277 Controls which reasoning items are rendered back to the model on later turns.

5278 If omitted or set to `auto`, the model determines the context mode. The

5279 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

5280 `current_turn`.

5281 

5276 When returned on a response, this is the effective reasoning context mode5282 When returned on a response, this is the effective reasoning context mode

5277 used for the response.5283 used for the response.

5278 5284 

5466 5472 

5467 Constrains the verbosity of the model's response. Lower values will result in5473 Constrains the verbosity of the model's response. Lower values will result in

5468 more concise responses, while higher values will result in more verbose responses.5474 more concise responses, while higher values will result in more verbose responses.

5469 Currently supported values are `low`, `medium`, and `high`.5475 Currently supported values are `low`, `medium`, and `high`. The default is

5476 `medium`.

5470 5477 

5471 - `"low"`5478 - `"low"`

5472 5479 

245 245 

246 An error object returned when the model fails to generate a Response.246 An error object returned when the model fails to generate a Response.

247 247 

248 - `code: "server_error" or "rate_limit_exceeded" or "invalid_prompt" or 16 more`248 - `code: "server_error" or "rate_limit_exceeded" or "invalid_prompt" or 17 more`

249 249 

250 The error code for the response.250 The error code for the response.

251 251 

255 255 

256 - `"invalid_prompt"`256 - `"invalid_prompt"`

257 257 

258 - `"data_residency_mismatch"`

259 

258 - `"bio_policy"`260 - `"bio_policy"`

259 261 

260 - `"vector_store_timeout"`262 - `"vector_store_timeout"`

5488 - `context: optional "auto" or "current_turn" or "all_turns"`5490 - `context: optional "auto" or "current_turn" or "all_turns"`

5489 5491 

5490 Controls which reasoning items are rendered back to the model on later turns.5492 Controls which reasoning items are rendered back to the model on later turns.

5493 If omitted or set to `auto`, the model determines the context mode. The

5494 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

5495 `current_turn`.

5496 

5491 When returned on a response, this is the effective reasoning context mode5497 When returned on a response, this is the effective reasoning context mode

5492 used for the response.5498 used for the response.

5493 5499 

5681 5687 

5682 Constrains the verbosity of the model's response. Lower values will result in5688 Constrains the verbosity of the model's response. Lower values will result in

5683 more concise responses, while higher values will result in more verbose responses.5689 more concise responses, while higher values will result in more verbose responses.

5684 Currently supported values are `low`, `medium`, and `high`.5690 Currently supported values are `low`, `medium`, and `high`. The default is

5691 `medium`.

5685 5692 

5686 - `"low"`5693 - `"low"`

5687 5694 

47 47 

48 An error object returned when the model fails to generate a Response.48 An error object returned when the model fails to generate a Response.

49 49 

50 - `code: "server_error" or "rate_limit_exceeded" or "invalid_prompt" or 16 more`50 - `code: "server_error" or "rate_limit_exceeded" or "invalid_prompt" or 17 more`

51 51 

52 The error code for the response.52 The error code for the response.

53 53 

57 57 

58 - `"invalid_prompt"`58 - `"invalid_prompt"`

59 59 

60 - `"data_residency_mismatch"`

61 

60 - `"bio_policy"`62 - `"bio_policy"`

61 63 

62 - `"vector_store_timeout"`64 - `"vector_store_timeout"`

5290 - `context: optional "auto" or "current_turn" or "all_turns"`5292 - `context: optional "auto" or "current_turn" or "all_turns"`

5291 5293 

5292 Controls which reasoning items are rendered back to the model on later turns.5294 Controls which reasoning items are rendered back to the model on later turns.

5295 If omitted or set to `auto`, the model determines the context mode. The

5296 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

5297 `current_turn`.

5298 

5293 When returned on a response, this is the effective reasoning context mode5299 When returned on a response, this is the effective reasoning context mode

5294 used for the response.5300 used for the response.

5295 5301 

5483 5489 

5484 Constrains the verbosity of the model's response. Lower values will result in5490 Constrains the verbosity of the model's response. Lower values will result in

5485 more concise responses, while higher values will result in more verbose responses.5491 more concise responses, while higher values will result in more verbose responses.

5486 Currently supported values are `low`, `medium`, and `high`.5492 Currently supported values are `low`, `medium`, and `high`. The default is

5493 `medium`.

5487 5494 

5488 - `"low"`5495 - `"low"`

5489 5496 

609 - `Context ReasoningContext`609 - `Context ReasoningContext`

610 610 

611 Controls which reasoning items are rendered back to the model on later turns.611 Controls which reasoning items are rendered back to the model on later turns.

612 If omitted or set to `auto`, the model determines the context mode. The

613 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

614 `current_turn`.

615 

612 When returned on a response, this is the effective reasoning context mode616 When returned on a response, this is the effective reasoning context mode

613 used for the response.617 used for the response.

614 618 

10 10 

11- `projectID string`11- `projectID string`

12 12 

13 The ID of the project.

14 

13- `serviceAccountID string`15- `serviceAccountID string`

14 16 

17 The ID of the service account.

18 

15- `body AdminOrganizationProjectServiceAccountAPIKeyNewParams`19- `body AdminOrganizationProjectServiceAccountAPIKeyNewParams`

16 20 

17 - `Name param.Field[string]`21 - `Name param.Field[string]`

28 32 

29 - `ID string`33 - `ID string`

30 34 

35 The identifier of the API key.

36 

31 - `CreatedAt int64`37 - `CreatedAt int64`

32 38 

39 The Unix timestamp (in seconds) when the API key was created.

40 

33 - `Name string`41 - `Name string`

34 42 

43 The name of the API key.

44 

35 - `Object OrganizationProjectServiceAccountAPIKey`45 - `Object OrganizationProjectServiceAccountAPIKey`

36 46 

37 The object type, which is always `organization.project.service_account.api_key`47 The object type, which is always `organization.project.service_account.api_key`

40 50 

41 - `Value string`51 - `Value string`

42 52 

53 The unredacted API key value.

54 

43### Example55### Example

44 56 

45```go57```go

1## Delete project spend limit

2 

3`client.Admin.Organization.Projects.SpendLimit.Delete(ctx, projectID) (*ProjectSpendLimitDeleted, error)`

4 

5**delete** `/organization/projects/{project_id}/spend_limit`

6 

7Delete a project's hard spend limit.

8 

9### Parameters

10 

11- `projectID string`

12 

13### Returns

14 

15- `type ProjectSpendLimitDeleted struct{…}`

16 

17 Confirmation payload returned after deleting a project hard spend limit.

18 

19 - `Deleted bool`

20 

21 Whether the hard spend limit was deleted.

22 

23 - `Object ProjectSpendLimitDeleted`

24 

25 The object type, which is always `project.spend_limit.deleted`.

26 

27 - `const ProjectSpendLimitDeletedProjectSpendLimitDeleted ProjectSpendLimitDeleted = "project.spend_limit.deleted"`

28 

29### Example

30 

31```go

32package main

33 

34import (

35 "context"

36 "fmt"

37 

38 "github.com/openai/openai-go"

39 "github.com/openai/openai-go/option"

40)

41 

42func main() {

43 client := openai.NewClient(

44 option.WithAdminAPIKey("My Admin API Key"),

45 )

46 projectSpendLimitDeleted, err := client.Admin.Organization.Projects.SpendLimit.Delete(context.TODO(), "proj_123")

47 if err != nil {

48 panic(err.Error())

49 }

50 fmt.Printf("%+v\n", projectSpendLimitDeleted.Deleted)

51}

52```

53 

54#### Response

55 

56```json

57{

58 "deleted": true,

59 "object": "project.spend_limit.deleted"

60}

61```

1## Retrieve project spend limit

2 

3`client.Admin.Organization.Projects.SpendLimit.Get(ctx, projectID) (*ProjectSpendLimit, error)`

4 

5**get** `/organization/projects/{project_id}/spend_limit`

6 

7Get a project's hard spend limit.

8 

9### Parameters

10 

11- `projectID string`

12 

13### Returns

14 

15- `type ProjectSpendLimit struct{…}`

16 

17 Represents a hard spend limit configured at the project level.

18 

19 - `Currency ProjectSpendLimitCurrency`

20 

21 The currency for the threshold amount. Currently, only `USD` is supported.

22 

23 - `string`

24 

25 - `type ProjectSpendLimitCurrency string`

26 

27 The currency for the threshold amount. Currently, only `USD` is supported.

28 

29 - `const ProjectSpendLimitCurrencyUsd ProjectSpendLimitCurrency = "USD"`

30 

31 - `Enforcement ProjectSpendLimitEnforcement`

32 

33 The current enforcement state of the hard spend limit.

34 

35 - `Status string`

36 

37 Whether the hard spend limit is currently enforcing.

38 

39 - `string`

40 

41 - `string`

42 

43 - `const ProjectSpendLimitEnforcementStatusInactive ProjectSpendLimitEnforcementStatus = "inactive"`

44 

45 - `const ProjectSpendLimitEnforcementStatusEnforcing ProjectSpendLimitEnforcementStatus = "enforcing"`

46 

47 - `Interval ProjectSpendLimitInterval`

48 

49 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

50 

51 - `string`

52 

53 - `type ProjectSpendLimitInterval string`

54 

55 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

56 

57 - `const ProjectSpendLimitIntervalMonth ProjectSpendLimitInterval = "month"`

58 

59 - `Object ProjectSpendLimit`

60 

61 The object type, which is always `project.spend_limit`.

62 

63 - `const ProjectSpendLimitProjectSpendLimit ProjectSpendLimit = "project.spend_limit"`

64 

65 - `ThresholdAmount int64`

66 

67 The hard spend limit amount, in cents.

68 

69### Example

70 

71```go

72package main

73 

74import (

75 "context"

76 "fmt"

77 

78 "github.com/openai/openai-go"

79 "github.com/openai/openai-go/option"

80)

81 

82func main() {

83 client := openai.NewClient(

84 option.WithAdminAPIKey("My Admin API Key"),

85 )

86 projectSpendLimit, err := client.Admin.Organization.Projects.SpendLimit.Get(context.TODO(), "proj_123")

87 if err != nil {

88 panic(err.Error())

89 }

90 fmt.Printf("%+v\n", projectSpendLimit.Currency)

91}

92```

93 

94#### Response

95 

96```json

97{

98 "currency": "USD",

99 "enforcement": {

100 "status": "inactive"

101 },

102 "interval": "month",

103 "object": "project.spend_limit",

104 "threshold_amount": 0

105}

106```

1## Update project spend limit

2 

3`client.Admin.Organization.Projects.SpendLimit.Update(ctx, projectID, body) (*ProjectSpendLimit, error)`

4 

5**post** `/organization/projects/{project_id}/spend_limit`

6 

7Create or replace a project's hard spend limit.

8 

9### Parameters

10 

11- `projectID string`

12 

13- `body AdminOrganizationProjectSpendLimitUpdateParams`

14 

15 - `Currency param.Field[AdminOrganizationProjectSpendLimitUpdateParamsCurrency]`

16 

17 The currency for the threshold amount. Currently, only `USD` is supported.

18 

19 - `const AdminOrganizationProjectSpendLimitUpdateParamsCurrencyUsd AdminOrganizationProjectSpendLimitUpdateParamsCurrency = "USD"`

20 

21 - `Interval param.Field[AdminOrganizationProjectSpendLimitUpdateParamsInterval]`

22 

23 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

24 

25 - `const AdminOrganizationProjectSpendLimitUpdateParamsIntervalMonth AdminOrganizationProjectSpendLimitUpdateParamsInterval = "month"`

26 

27 - `ThresholdAmount param.Field[int64]`

28 

29 The hard spend limit amount, in cents.

30 

31### Returns

32 

33- `type ProjectSpendLimit struct{…}`

34 

35 Represents a hard spend limit configured at the project level.

36 

37 - `Currency ProjectSpendLimitCurrency`

38 

39 The currency for the threshold amount. Currently, only `USD` is supported.

40 

41 - `string`

42 

43 - `type ProjectSpendLimitCurrency string`

44 

45 The currency for the threshold amount. Currently, only `USD` is supported.

46 

47 - `const ProjectSpendLimitCurrencyUsd ProjectSpendLimitCurrency = "USD"`

48 

49 - `Enforcement ProjectSpendLimitEnforcement`

50 

51 The current enforcement state of the hard spend limit.

52 

53 - `Status string`

54 

55 Whether the hard spend limit is currently enforcing.

56 

57 - `string`

58 

59 - `string`

60 

61 - `const ProjectSpendLimitEnforcementStatusInactive ProjectSpendLimitEnforcementStatus = "inactive"`

62 

63 - `const ProjectSpendLimitEnforcementStatusEnforcing ProjectSpendLimitEnforcementStatus = "enforcing"`

64 

65 - `Interval ProjectSpendLimitInterval`

66 

67 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

68 

69 - `string`

70 

71 - `type ProjectSpendLimitInterval string`

72 

73 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

74 

75 - `const ProjectSpendLimitIntervalMonth ProjectSpendLimitInterval = "month"`

76 

77 - `Object ProjectSpendLimit`

78 

79 The object type, which is always `project.spend_limit`.

80 

81 - `const ProjectSpendLimitProjectSpendLimit ProjectSpendLimit = "project.spend_limit"`

82 

83 - `ThresholdAmount int64`

84 

85 The hard spend limit amount, in cents.

86 

87### Example

88 

89```go

90package main

91 

92import (

93 "context"

94 "fmt"

95 

96 "github.com/openai/openai-go"

97 "github.com/openai/openai-go/option"

98)

99 

100func main() {

101 client := openai.NewClient(

102 option.WithAdminAPIKey("My Admin API Key"),

103 )

104 projectSpendLimit, err := client.Admin.Organization.Projects.SpendLimit.Update(

105 context.TODO(),

106 "proj_123",

107 openai.AdminOrganizationProjectSpendLimitUpdateParams{

108 Currency: openai.AdminOrganizationProjectSpendLimitUpdateParamsCurrencyUsd,

109 Interval: openai.AdminOrganizationProjectSpendLimitUpdateParamsIntervalMonth,

110 ThresholdAmount: 1,

111 },

112 )

113 if err != nil {

114 panic(err.Error())

115 }

116 fmt.Printf("%+v\n", projectSpendLimit.Currency)

117}

118```

119 

120#### Response

121 

122```json

123{

124 "currency": "USD",

125 "enforcement": {

126 "status": "inactive"

127 },

128 "interval": "month",

129 "object": "project.spend_limit",

130 "threshold_amount": 0

131}

132```

1## Delete organization spend limit

2 

3`client.Admin.Organization.SpendLimit.Delete(ctx) (*OrganizationSpendLimitDeleted, error)`

4 

5**delete** `/organization/spend_limit`

6 

7Delete the organization's hard spend limit.

8 

9### Returns

10 

11- `type OrganizationSpendLimitDeleted struct{…}`

12 

13 Confirmation payload returned after deleting an organization hard spend limit.

14 

15 - `Deleted bool`

16 

17 Whether the hard spend limit was deleted.

18 

19 - `Object OrganizationSpendLimitDeleted`

20 

21 The object type, which is always `organization.spend_limit.deleted`.

22 

23 - `const OrganizationSpendLimitDeletedOrganizationSpendLimitDeleted OrganizationSpendLimitDeleted = "organization.spend_limit.deleted"`

24 

25### Example

26 

27```go

28package main

29 

30import (

31 "context"

32 "fmt"

33 

34 "github.com/openai/openai-go"

35 "github.com/openai/openai-go/option"

36)

37 

38func main() {

39 client := openai.NewClient(

40 option.WithAdminAPIKey("My Admin API Key"),

41 )

42 organizationSpendLimitDeleted, err := client.Admin.Organization.SpendLimit.Delete(context.TODO())

43 if err != nil {

44 panic(err.Error())

45 }

46 fmt.Printf("%+v\n", organizationSpendLimitDeleted.Deleted)

47}

48```

49 

50#### Response

51 

52```json

53{

54 "deleted": true,

55 "object": "organization.spend_limit.deleted"

56}

57```

1## Retrieve organization spend limit

2 

3`client.Admin.Organization.SpendLimit.Get(ctx) (*OrganizationSpendLimit, error)`

4 

5**get** `/organization/spend_limit`

6 

7Get the organization's hard spend limit.

8 

9### Returns

10 

11- `type OrganizationSpendLimit struct{…}`

12 

13 Represents a hard spend limit configured at the organization level.

14 

15 - `Currency OrganizationSpendLimitCurrency`

16 

17 The currency for the threshold amount. Currently, only `USD` is supported.

18 

19 - `string`

20 

21 - `type OrganizationSpendLimitCurrency string`

22 

23 The currency for the threshold amount. Currently, only `USD` is supported.

24 

25 - `const OrganizationSpendLimitCurrencyUsd OrganizationSpendLimitCurrency = "USD"`

26 

27 - `Enforcement OrganizationSpendLimitEnforcement`

28 

29 The current enforcement state of the hard spend limit.

30 

31 - `Status string`

32 

33 Whether the hard spend limit is currently enforcing.

34 

35 - `string`

36 

37 - `string`

38 

39 - `const OrganizationSpendLimitEnforcementStatusInactive OrganizationSpendLimitEnforcementStatus = "inactive"`

40 

41 - `const OrganizationSpendLimitEnforcementStatusEnforcing OrganizationSpendLimitEnforcementStatus = "enforcing"`

42 

43 - `Interval OrganizationSpendLimitInterval`

44 

45 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

46 

47 - `string`

48 

49 - `type OrganizationSpendLimitInterval string`

50 

51 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

52 

53 - `const OrganizationSpendLimitIntervalMonth OrganizationSpendLimitInterval = "month"`

54 

55 - `Object OrganizationSpendLimit`

56 

57 The object type, which is always `organization.spend_limit`.

58 

59 - `const OrganizationSpendLimitOrganizationSpendLimit OrganizationSpendLimit = "organization.spend_limit"`

60 

61 - `ThresholdAmount int64`

62 

63 The hard spend limit amount, in cents.

64 

65### Example

66 

67```go

68package main

69 

70import (

71 "context"

72 "fmt"

73 

74 "github.com/openai/openai-go"

75 "github.com/openai/openai-go/option"

76)

77 

78func main() {

79 client := openai.NewClient(

80 option.WithAdminAPIKey("My Admin API Key"),

81 )

82 organizationSpendLimit, err := client.Admin.Organization.SpendLimit.Get(context.TODO())

83 if err != nil {

84 panic(err.Error())

85 }

86 fmt.Printf("%+v\n", organizationSpendLimit.Currency)

87}

88```

89 

90#### Response

91 

92```json

93{

94 "currency": "USD",

95 "enforcement": {

96 "status": "inactive"

97 },

98 "interval": "month",

99 "object": "organization.spend_limit",

100 "threshold_amount": 0

101}

102```

1## Update organization spend limit

2 

3`client.Admin.Organization.SpendLimit.Update(ctx, body) (*OrganizationSpendLimit, error)`

4 

5**post** `/organization/spend_limit`

6 

7Create or replace the organization's hard spend limit.

8 

9### Parameters

10 

11- `body AdminOrganizationSpendLimitUpdateParams`

12 

13 - `Currency param.Field[AdminOrganizationSpendLimitUpdateParamsCurrency]`

14 

15 The currency for the threshold amount. Currently, only `USD` is supported.

16 

17 - `const AdminOrganizationSpendLimitUpdateParamsCurrencyUsd AdminOrganizationSpendLimitUpdateParamsCurrency = "USD"`

18 

19 - `Interval param.Field[AdminOrganizationSpendLimitUpdateParamsInterval]`

20 

21 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

22 

23 - `const AdminOrganizationSpendLimitUpdateParamsIntervalMonth AdminOrganizationSpendLimitUpdateParamsInterval = "month"`

24 

25 - `ThresholdAmount param.Field[int64]`

26 

27 The hard spend limit amount, in cents.

28 

29### Returns

30 

31- `type OrganizationSpendLimit struct{…}`

32 

33 Represents a hard spend limit configured at the organization level.

34 

35 - `Currency OrganizationSpendLimitCurrency`

36 

37 The currency for the threshold amount. Currently, only `USD` is supported.

38 

39 - `string`

40 

41 - `type OrganizationSpendLimitCurrency string`

42 

43 The currency for the threshold amount. Currently, only `USD` is supported.

44 

45 - `const OrganizationSpendLimitCurrencyUsd OrganizationSpendLimitCurrency = "USD"`

46 

47 - `Enforcement OrganizationSpendLimitEnforcement`

48 

49 The current enforcement state of the hard spend limit.

50 

51 - `Status string`

52 

53 Whether the hard spend limit is currently enforcing.

54 

55 - `string`

56 

57 - `string`

58 

59 - `const OrganizationSpendLimitEnforcementStatusInactive OrganizationSpendLimitEnforcementStatus = "inactive"`

60 

61 - `const OrganizationSpendLimitEnforcementStatusEnforcing OrganizationSpendLimitEnforcementStatus = "enforcing"`

62 

63 - `Interval OrganizationSpendLimitInterval`

64 

65 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

66 

67 - `string`

68 

69 - `type OrganizationSpendLimitInterval string`

70 

71 The time interval for evaluating spend against the threshold. Currently, only `month` is supported.

72 

73 - `const OrganizationSpendLimitIntervalMonth OrganizationSpendLimitInterval = "month"`

74 

75 - `Object OrganizationSpendLimit`

76 

77 The object type, which is always `organization.spend_limit`.

78 

79 - `const OrganizationSpendLimitOrganizationSpendLimit OrganizationSpendLimit = "organization.spend_limit"`

80 

81 - `ThresholdAmount int64`

82 

83 The hard spend limit amount, in cents.

84 

85### Example

86 

87```go

88package main

89 

90import (

91 "context"

92 "fmt"

93 

94 "github.com/openai/openai-go"

95 "github.com/openai/openai-go/option"

96)

97 

98func main() {

99 client := openai.NewClient(

100 option.WithAdminAPIKey("My Admin API Key"),

101 )

102 organizationSpendLimit, err := client.Admin.Organization.SpendLimit.Update(context.TODO(), openai.AdminOrganizationSpendLimitUpdateParams{

103 Currency: openai.AdminOrganizationSpendLimitUpdateParamsCurrencyUsd,

104 Interval: openai.AdminOrganizationSpendLimitUpdateParamsIntervalMonth,

105 ThresholdAmount: 1,

106 })

107 if err != nil {

108 panic(err.Error())

109 }

110 fmt.Printf("%+v\n", organizationSpendLimit.Currency)

111}

112```

113 

114#### Response

115 

116```json

117{

118 "currency": "USD",

119 "enforcement": {

120 "status": "inactive"

121 },

122 "interval": "month",

123 "object": "organization.spend_limit",

124 "threshold_amount": 0

125}

126```

46 46 

47 - `const BetaResponseErrorCodeInvalidPrompt BetaResponseErrorCode = "invalid_prompt"`47 - `const BetaResponseErrorCodeInvalidPrompt BetaResponseErrorCode = "invalid_prompt"`

48 48 

49 - `const BetaResponseErrorCodeDataResidencyMismatch BetaResponseErrorCode = "data_residency_mismatch"`

50 

49 - `const BetaResponseErrorCodeBioPolicy BetaResponseErrorCode = "bio_policy"`51 - `const BetaResponseErrorCodeBioPolicy BetaResponseErrorCode = "bio_policy"`

50 52 

51 - `const BetaResponseErrorCodeVectorStoreTimeout BetaResponseErrorCode = "vector_store_timeout"`53 - `const BetaResponseErrorCodeVectorStoreTimeout BetaResponseErrorCode = "vector_store_timeout"`

6160 - `Context string`6162 - `Context string`

6161 6163 

6162 Controls which reasoning items are rendered back to the model on later turns.6164 Controls which reasoning items are rendered back to the model on later turns.

6165 If omitted or set to `auto`, the model determines the context mode. The

6166 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

6167 `current_turn`.

6168 

6163 When returned on a response, this is the effective reasoning context mode6169 When returned on a response, this is the effective reasoning context mode

6164 used for the response.6170 used for the response.

6165 6171 

6363 6369 

6364 Constrains the verbosity of the model's response. Lower values will result in6370 Constrains the verbosity of the model's response. Lower values will result in

6365 more concise responses, while higher values will result in more verbose responses.6371 more concise responses, while higher values will result in more verbose responses.

6366 Currently supported values are `low`, `medium`, and `high`.6372 Currently supported values are `low`, `medium`, and `high`. The default is

6373 `medium`.

6367 6374 

6368 - `const BetaResponseTextConfigVerbosityLow BetaResponseTextConfigVerbosity = "low"`6375 - `const BetaResponseTextConfigVerbosityLow BetaResponseTextConfigVerbosity = "low"`

6369 6376 

4220 - `Context string`4220 - `Context string`

4221 4221 

4222 Controls which reasoning items are rendered back to the model on later turns.4222 Controls which reasoning items are rendered back to the model on later turns.

4223 If omitted or set to `auto`, the model determines the context mode. The

4224 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

4225 `current_turn`.

4226 

4223 When returned on a response, this is the effective reasoning context mode4227 When returned on a response, this is the effective reasoning context mode

4224 used for the response.4228 used for the response.

4225 4229 

4677 4681 

4678 - `const BetaResponseErrorCodeInvalidPrompt BetaResponseErrorCode = "invalid_prompt"`4682 - `const BetaResponseErrorCodeInvalidPrompt BetaResponseErrorCode = "invalid_prompt"`

4679 4683 

4684 - `const BetaResponseErrorCodeDataResidencyMismatch BetaResponseErrorCode = "data_residency_mismatch"`

4685 

4680 - `const BetaResponseErrorCodeBioPolicy BetaResponseErrorCode = "bio_policy"`4686 - `const BetaResponseErrorCodeBioPolicy BetaResponseErrorCode = "bio_policy"`

4681 4687 

4682 - `const BetaResponseErrorCodeVectorStoreTimeout BetaResponseErrorCode = "vector_store_timeout"`4688 - `const BetaResponseErrorCodeVectorStoreTimeout BetaResponseErrorCode = "vector_store_timeout"`

10791 - `Context string`10797 - `Context string`

10792 10798 

10793 Controls which reasoning items are rendered back to the model on later turns.10799 Controls which reasoning items are rendered back to the model on later turns.

10800 If omitted or set to `auto`, the model determines the context mode. The

10801 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

10802 `current_turn`.

10803 

10794 When returned on a response, this is the effective reasoning context mode10804 When returned on a response, this is the effective reasoning context mode

10795 used for the response.10805 used for the response.

10796 10806 

10994 11004 

10995 Constrains the verbosity of the model's response. Lower values will result in11005 Constrains the verbosity of the model's response. Lower values will result in

10996 more concise responses, while higher values will result in more verbose responses.11006 more concise responses, while higher values will result in more verbose responses.

10997 Currently supported values are `low`, `medium`, and `high`.11007 Currently supported values are `low`, `medium`, and `high`. The default is

11008 `medium`.

10998 11009 

10999 - `const BetaResponseTextConfigVerbosityLow BetaResponseTextConfigVerbosity = "low"`11010 - `const BetaResponseTextConfigVerbosityLow BetaResponseTextConfigVerbosity = "low"`

11000 11011 

81 81 

82 - `const BetaResponseErrorCodeInvalidPrompt BetaResponseErrorCode = "invalid_prompt"`82 - `const BetaResponseErrorCodeInvalidPrompt BetaResponseErrorCode = "invalid_prompt"`

83 83 

84 - `const BetaResponseErrorCodeDataResidencyMismatch BetaResponseErrorCode = "data_residency_mismatch"`

85 

84 - `const BetaResponseErrorCodeBioPolicy BetaResponseErrorCode = "bio_policy"`86 - `const BetaResponseErrorCodeBioPolicy BetaResponseErrorCode = "bio_policy"`

85 87 

86 - `const BetaResponseErrorCodeVectorStoreTimeout BetaResponseErrorCode = "vector_store_timeout"`88 - `const BetaResponseErrorCodeVectorStoreTimeout BetaResponseErrorCode = "vector_store_timeout"`

6195 - `Context string`6197 - `Context string`

6196 6198 

6197 Controls which reasoning items are rendered back to the model on later turns.6199 Controls which reasoning items are rendered back to the model on later turns.

6200 If omitted or set to `auto`, the model determines the context mode. The

6201 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

6202 `current_turn`.

6203 

6198 When returned on a response, this is the effective reasoning context mode6204 When returned on a response, this is the effective reasoning context mode

6199 used for the response.6205 used for the response.

6200 6206 

6398 6404 

6399 Constrains the verbosity of the model's response. Lower values will result in6405 Constrains the verbosity of the model's response. Lower values will result in

6400 more concise responses, while higher values will result in more verbose responses.6406 more concise responses, while higher values will result in more verbose responses.

6401 Currently supported values are `low`, `medium`, and `high`.6407 Currently supported values are `low`, `medium`, and `high`. The default is

6408 `medium`.

6402 6409 

6403 - `const BetaResponseTextConfigVerbosityLow BetaResponseTextConfigVerbosity = "low"`6410 - `const BetaResponseTextConfigVerbosityLow BetaResponseTextConfigVerbosity = "low"`

6404 6411 

3863 - `Context string`3863 - `Context string`

3864 3864 

3865 Controls which reasoning items are rendered back to the model on later turns.3865 Controls which reasoning items are rendered back to the model on later turns.

3866 If omitted or set to `auto`, the model determines the context mode. The

3867 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

3868 `current_turn`.

3869 

3866 When returned on a response, this is the effective reasoning context mode3870 When returned on a response, this is the effective reasoning context mode

3867 used for the response.3871 used for the response.

3868 3872 

4023 4027 

4024 Constrains the verbosity of the model's response. Lower values will result in4028 Constrains the verbosity of the model's response. Lower values will result in

4025 more concise responses, while higher values will result in more verbose responses.4029 more concise responses, while higher values will result in more verbose responses.

4026 Currently supported values are `low`, `medium`, and `high`.4030 Currently supported values are `low`, `medium`, and `high`. The default is

4031 `medium`.

4027 4032 

4028 - `const BetaResponseInputTokenCountParamsTextVerbosityLow BetaResponseInputTokenCountParamsTextVerbosity = "low"`4033 - `const BetaResponseInputTokenCountParamsTextVerbosityLow BetaResponseInputTokenCountParamsTextVerbosity = "low"`

4029 4034 

1085 1085 

1086 Constrains the verbosity of the model's response. Lower values will result in1086 Constrains the verbosity of the model's response. Lower values will result in

1087 more concise responses, while higher values will result in more verbose responses.1087 more concise responses, while higher values will result in more verbose responses.

1088 Currently supported values are `low`, `medium`, and `high`.1088 Currently supported values are `low`, `medium`, and `high`. The default is

1089 `medium`.

1089 1090 

1090 - `const ChatCompletionNewParamsVerbosityLow ChatCompletionNewParamsVerbosity = "low"`1091 - `const ChatCompletionNewParamsVerbosityLow ChatCompletionNewParamsVerbosity = "low"`

1091 1092 

38 38 

39 - `const ResponseErrorCodeInvalidPrompt ResponseErrorCode = "invalid_prompt"`39 - `const ResponseErrorCodeInvalidPrompt ResponseErrorCode = "invalid_prompt"`

40 40 

41 - `const ResponseErrorCodeDataResidencyMismatch ResponseErrorCode = "data_residency_mismatch"`

42 

41 - `const ResponseErrorCodeBioPolicy ResponseErrorCode = "bio_policy"`43 - `const ResponseErrorCodeBioPolicy ResponseErrorCode = "bio_policy"`

42 44 

43 - `const ResponseErrorCodeVectorStoreTimeout ResponseErrorCode = "vector_store_timeout"`45 - `const ResponseErrorCodeVectorStoreTimeout ResponseErrorCode = "vector_store_timeout"`

5379 - `Context ReasoningContext`5381 - `Context ReasoningContext`

5380 5382 

5381 Controls which reasoning items are rendered back to the model on later turns.5383 Controls which reasoning items are rendered back to the model on later turns.

5384 If omitted or set to `auto`, the model determines the context mode. The

5385 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

5386 `current_turn`.

5387 

5382 When returned on a response, this is the effective reasoning context mode5388 When returned on a response, this is the effective reasoning context mode

5383 used for the response.5389 used for the response.

5384 5390 

5586 5592 

5587 Constrains the verbosity of the model's response. Lower values will result in5593 Constrains the verbosity of the model's response. Lower values will result in

5588 more concise responses, while higher values will result in more verbose responses.5594 more concise responses, while higher values will result in more verbose responses.

5589 Currently supported values are `low`, `medium`, and `high`.5595 Currently supported values are `low`, `medium`, and `high`. The default is

5596 `medium`.

5590 5597 

5591 - `const ResponseTextConfigVerbosityLow ResponseTextConfigVerbosity = "low"`5598 - `const ResponseTextConfigVerbosityLow ResponseTextConfigVerbosity = "low"`

5592 5599 

4014 4014 

4015 - `const ResponseErrorCodeInvalidPrompt ResponseErrorCode = "invalid_prompt"`4015 - `const ResponseErrorCodeInvalidPrompt ResponseErrorCode = "invalid_prompt"`

4016 4016 

4017 - `const ResponseErrorCodeDataResidencyMismatch ResponseErrorCode = "data_residency_mismatch"`

4018 

4017 - `const ResponseErrorCodeBioPolicy ResponseErrorCode = "bio_policy"`4019 - `const ResponseErrorCodeBioPolicy ResponseErrorCode = "bio_policy"`

4018 4020 

4019 - `const ResponseErrorCodeVectorStoreTimeout ResponseErrorCode = "vector_store_timeout"`4021 - `const ResponseErrorCodeVectorStoreTimeout ResponseErrorCode = "vector_store_timeout"`

9355 - `Context ReasoningContext`9357 - `Context ReasoningContext`

9356 9358 

9357 Controls which reasoning items are rendered back to the model on later turns.9359 Controls which reasoning items are rendered back to the model on later turns.

9360 If omitted or set to `auto`, the model determines the context mode. The

9361 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

9362 `current_turn`.

9363 

9358 When returned on a response, this is the effective reasoning context mode9364 When returned on a response, this is the effective reasoning context mode

9359 used for the response.9365 used for the response.

9360 9366 

9562 9568 

9563 Constrains the verbosity of the model's response. Lower values will result in9569 Constrains the verbosity of the model's response. Lower values will result in

9564 more concise responses, while higher values will result in more verbose responses.9570 more concise responses, while higher values will result in more verbose responses.

9565 Currently supported values are `low`, `medium`, and `high`.9571 Currently supported values are `low`, `medium`, and `high`. The default is

9572 `medium`.

9566 9573 

9567 - `const ResponseTextConfigVerbosityLow ResponseTextConfigVerbosity = "low"`9574 - `const ResponseTextConfigVerbosityLow ResponseTextConfigVerbosity = "low"`

9568 9575 

75 75 

76 - `const ResponseErrorCodeInvalidPrompt ResponseErrorCode = "invalid_prompt"`76 - `const ResponseErrorCodeInvalidPrompt ResponseErrorCode = "invalid_prompt"`

77 77 

78 - `const ResponseErrorCodeDataResidencyMismatch ResponseErrorCode = "data_residency_mismatch"`

79 

78 - `const ResponseErrorCodeBioPolicy ResponseErrorCode = "bio_policy"`80 - `const ResponseErrorCodeBioPolicy ResponseErrorCode = "bio_policy"`

79 81 

80 - `const ResponseErrorCodeVectorStoreTimeout ResponseErrorCode = "vector_store_timeout"`82 - `const ResponseErrorCodeVectorStoreTimeout ResponseErrorCode = "vector_store_timeout"`

5416 - `Context ReasoningContext`5418 - `Context ReasoningContext`

5417 5419 

5418 Controls which reasoning items are rendered back to the model on later turns.5420 Controls which reasoning items are rendered back to the model on later turns.

5421 If omitted or set to `auto`, the model determines the context mode. The

5422 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

5423 `current_turn`.

5424 

5419 When returned on a response, this is the effective reasoning context mode5425 When returned on a response, this is the effective reasoning context mode

5420 used for the response.5426 used for the response.

5421 5427 

5623 5629 

5624 Constrains the verbosity of the model's response. Lower values will result in5630 Constrains the verbosity of the model's response. Lower values will result in

5625 more concise responses, while higher values will result in more verbose responses.5631 more concise responses, while higher values will result in more verbose responses.

5626 Currently supported values are `low`, `medium`, and `high`.5632 Currently supported values are `low`, `medium`, and `high`. The default is

5633 `medium`.

5627 5634 

5628 - `const ResponseTextConfigVerbosityLow ResponseTextConfigVerbosity = "low"`5635 - `const ResponseTextConfigVerbosityLow ResponseTextConfigVerbosity = "low"`

5629 5636 

3573 3573 

3574 Constrains the verbosity of the model's response. Lower values will result in3574 Constrains the verbosity of the model's response. Lower values will result in

3575 more concise responses, while higher values will result in more verbose responses.3575 more concise responses, while higher values will result in more verbose responses.

3576 Currently supported values are `low`, `medium`, and `high`.3576 Currently supported values are `low`, `medium`, and `high`. The default is

3577 `medium`.

3577 3578 

3578 - `const InputTokenCountParamsTextVerbosityLow InputTokenCountParamsTextVerbosity = "low"`3579 - `const InputTokenCountParamsTextVerbosityLow InputTokenCountParamsTextVerbosity = "low"`

3579 3580 

592 - `Optional<Context> context`592 - `Optional<Context> context`

593 593 

594 Controls which reasoning items are rendered back to the model on later turns.594 Controls which reasoning items are rendered back to the model on later turns.

595 If omitted or set to `auto`, the model determines the context mode. The

596 `gpt-5.6` model family defaults to `all_turns`; earlier models default to

597 `current_turn`.

598 

595 When returned on a response, this is the effective reasoning context mode599 When returned on a response, this is the effective reasoning context mode

596 used for the response.600 used for the response.

597 601 

12 12 

13 - `String projectId`13 - `String projectId`

14 14 

15 The ID of the project.

16 

15 - `Optional<String> serviceAccountId`17 - `Optional<String> serviceAccountId`

16 18 

19 The ID of the service account.

20 

17 - `Optional<String> name`21 - `Optional<String> name`

18 22 

19 API key name.23 API key name.

28 32 

29 - `String id`33 - `String id`

30 34 

35 The identifier of the API key.

36 

31 - `long createdAt`37 - `long createdAt`

32 38 

39 The Unix timestamp (in seconds) when the API key was created.

40 

33 - `String name`41 - `String name`

34 42 

43 The name of the API key.

44 

35 - `JsonValue; object_ "organization.project.service_account.api_key"constant`45 - `JsonValue; object_ "organization.project.service_account.api_key"constant`

36 46 

37 The object type, which is always `organization.project.service_account.api_key`47 The object type, which is always `organization.project.service_account.api_key`

40 50 

41 - `String value`51 - `String value`

42 52 

53 The unredacted API key value.

54 

43### Example55### Example

44 56 

45```java57```java

1## Delete project spend limit

2 

3`ProjectSpendLimitDeleted admin().organization().projects().spendLimit().delete(SpendLimitDeleteParamsparams = SpendLimitDeleteParams.none(), RequestOptionsrequestOptions = RequestOptions.none())`

4 

5**delete** `/organization/projects/{project_id}/spend_limit`

6 

7Delete a project's hard spend limit.

8 

9### Parameters

10 

11- `SpendLimitDeleteParams params`

12 

13 - `Optional<String> projectId`

14 

15### Returns

16 

17- `class ProjectSpendLimitDeleted:`

18 

19 Confirmation payload returned after deleting a project hard spend limit.

20 

21 - `boolean deleted`

22 

23 Whether the hard spend limit was deleted.

24 

25 - `JsonValue; object_ "project.spend_limit.deleted"constant`

26 

27 The object type, which is always `project.spend_limit.deleted`.

28 

29 - `PROJECT_SPEND_LIMIT_DELETED("project.spend_limit.deleted")`

30 

31### Example

32 

33```java

34package com.openai.example;

35 

36import com.openai.client.OpenAIClient;

37import com.openai.client.okhttp.OpenAIOkHttpClient;

38import com.openai.models.admin.organization.projects.spendlimit.ProjectSpendLimitDeleted;

39import com.openai.models.admin.organization.projects.spendlimit.SpendLimitDeleteParams;

40 

41public final class Main {

42 private Main() {}

43 

44 public static void main(String[] args) {

45 OpenAIClient client = OpenAIOkHttpClient.fromEnv();

46 

47 ProjectSpendLimitDeleted projectSpendLimitDeleted = client.admin().organization().projects().spendLimit().delete("proj_123");

48 }

49}

50```

51 

52#### Response

53 

54```json

55{

56 "deleted": true,

57 "object": "project.spend_limit.deleted"

58}

59```

1## Retrieve project spend limit

2 

3`ProjectSpendLimit admin().organization().projects().spendLimit().retrieve(SpendLimitRetrieveParamsparams = SpendLimitRetrieveParams.none(), RequestOptionsrequestOptions = RequestOptions.none())`

4 

5**get** `/organization/projects/{project_id}/spend_limit`

6 

7Get a project's hard spend limit.

8 

9### Parameters

10 

11- `SpendLimitRetrieveParams params`

12 

13 - `Optional<String> projectId`

14 

15### Returns

16 

17- `class ProjectSpendLimit:`

18 

19 Represents a hard spend limit configured at the project level.

20 

21 - `Currency currency`

22 

23 The currency for the threshold amount. Currently, only `USD` is supported.

24 

25 - `USD("USD")`

26 

27 - `Enforcement enforcement`

28 

29 The current enforcement state of the hard spend limit.

30 

31 - `Status status`