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

30<!-- x-release-please-start-version -->30<!-- x-release-please-start-version -->

31 31 

32```sh32```sh

33go get -u 'github.com/openai/openai-go/v3@v3.47.0'33go get -u 'github.com/openai/openai-go/v3@v3.48.0'

34```34```

35 35 

36<!-- x-release-please-end -->36<!-- x-release-please-end -->

958accepted (this overwrites any previous client) and receives requests after any958accepted (this overwrites any previous client) and receives requests after any

959middleware has been applied.959middleware has been applied.

960 960 

961### Mutual TLS with a custom HTTP client

962 

963For API-key authenticated HTTP requests that require mutual TLS, configure a

964native Go `*http.Client` and pass it through `option.WithHTTPClient`. The

965certificate file must contain the client leaf followed by every required

966intermediate. Presenting intermediates requires certificate-chain support to be

967enabled for your organization; otherwise, the client certificate must be

968signed directly by an active uploaded certificate. See the

969[OpenAI mTLS setup requirements](https://help.openai.com/en/articles/10876024):

970 

971```go

972certificate, err := tls.LoadX509KeyPair(

973 "/secrets/openai/client-chain.pem",

974 "/secrets/openai/client.key",

975)

976if err != nil {

977 return err

978}

979 

980defaultTransport, ok := http.DefaultTransport.(*http.Transport)

981if !ok {

982 return errors.New("http.DefaultTransport is not an *http.Transport")

983}

984transport := defaultTransport.Clone()

985transport.Proxy = nil

986transport.DialTLS = nil

987transport.DialTLSContext = nil

988transport.ResponseHeaderTimeout = 10 * time.Minute

989transport.TLSClientConfig = &tls.Config{

990 Certificates: []tls.Certificate{certificate},

991 GetClientCertificate: func(*tls.CertificateRequestInfo) (*tls.Certificate, error) {

992 return &certificate, nil

993 },

994}

995 

996httpClient := &http.Client{

997 Transport: transport,

998 CheckRedirect: func(*http.Request, []*http.Request) error {

999 return http.ErrUseLastResponse

1000 },

1001}

1002 

1003client := openai.NewClient(

1004 option.WithBaseURL("https://mtls.api.openai.com/v1"),

1005 option.WithHTTPClient(httpClient),

1006)

1007 

1008if _, err := client.Models.List(context.Background()); err != nil {

1009 return err

1010}

1011```

1012 

1013The SDK does not select an mTLS endpoint automatically when a custom HTTP

1014client is used. The explicit `option.WithBaseURL` above overrides

1015`OPENAI_BASE_URL`; replace it with `https://mtls-eu.api.openai.com/v1` for the

1016EU endpoint, or remove it to use `OPENAI_BASE_URL`. Keep server trust separate

1017by configuring `RootCAs` on the fresh `tls.Config` when custom roots are

1018required.

1019 

1020`tls.LoadX509KeyPair` fails for unreadable files and for malformed or mismatched

1021leaf/key material. It loads later `CERTIFICATE` blocks into the presented chain

1022without validating those intermediates. Certificate validity, intermediate

1023parsing, chain trust, and OpenAI product policy remain TLS-handshake/server

1024checks. Rebuild the transport and OpenAI client after rotating a certificate

1025because existing TLS connections cannot renegotiate client authentication.

1026When overriding the HTTP client, the application also owns redirect, proxy, and

1027timeout policy. This dedicated client bypasses proxies, retains the SDK's

102810-minute response-header timeout, replaces inherited client-certificate

1029callbacks, TLS dial hooks, and TLS session state with a fresh TLS config, and

1030disables redirects so the client certificate is only offered to the configured

1031API endpoint. Its callback always returns the configured certificate because

1032Go's automatic selection can otherwise suppress it when a server's

1033acceptable-CA hint does not match the local chain. If a proxy is required, use

1034a transport that keeps the proxy TLS configuration separate from the origin

1035client certificate.

1036 

1037The complete tested recipe is in

1038[`examples/mutual-tls`](./examples/mutual-tls/main.go).

1039 

961## Workload Identity Authentication1040## Workload Identity Authentication

962 1041 

963For cloud workloads (Kubernetes, Azure, Google Cloud Platform), you can use workload identity authentication instead of API keys. This provides short-lived tokens that are automatically refreshed.1042For cloud workloads (Kubernetes, Azure, Google Cloud Platform), you can use workload identity authentication instead of API keys. This provides short-lived tokens that are automatically refreshed.

2 2 

3<!-- x-release-please-start-version -->3<!-- x-release-please-start-version -->

4 4 

5[![Maven Central](https://img.shields.io/maven-central/v/com.openai/openai-java)](https://central.sonatype.com/artifact/com.openai/openai-java/4.46.0)5[![Maven Central](https://img.shields.io/maven-central/v/com.openai/openai-java)](https://central.sonatype.com/artifact/com.openai/openai-java/4.47.0)

6[![javadoc](https://javadoc.io/badge2/com.openai/openai-java/4.46.0/javadoc.svg)](https://javadoc.io/doc/com.openai/openai-java/4.46.0)6[![javadoc](https://javadoc.io/badge2/com.openai/openai-java/4.47.0/javadoc.svg)](https://javadoc.io/doc/com.openai/openai-java/4.47.0)

7 7 

8<!-- x-release-please-end -->8<!-- x-release-please-end -->

9 9 

11 11 

12<!-- x-release-please-start-version -->12<!-- x-release-please-start-version -->

13 13 

14The REST API documentation can be found on [platform.openai.com](https://platform.openai.com/docs). Javadocs are available on [javadoc.io](https://javadoc.io/doc/com.openai/openai-java/4.46.0).14The REST API documentation can be found on [platform.openai.com](https://platform.openai.com/docs). Javadocs are available on [javadoc.io](https://javadoc.io/doc/com.openai/openai-java/4.47.0).

15 15 

16<!-- x-release-please-end -->16<!-- x-release-please-end -->

17 17 

22### Gradle22### Gradle

23 23 

24```kotlin24```kotlin

25implementation("com.openai:openai-java:4.46.0")25implementation("com.openai:openai-java:4.47.0")

26```26```

27 27 

28### Maven28### Maven

31<dependency>31<dependency>

32 <groupId>com.openai</groupId>32 <groupId>com.openai</groupId>

33 <artifactId>openai-java</artifactId>33 <artifactId>openai-java</artifactId>

34 <version>4.46.0</version>34 <version>4.47.0</version>

35</dependency>35</dependency>

36```36```

37 37 

94<!-- x-release-please-start-version -->94<!-- x-release-please-start-version -->

95 95 

96```kotlin96```kotlin

97implementation("com.openai:openai-java-bedrock:4.46.0")97implementation("com.openai:openai-java-bedrock:4.47.0")

98```98```

99 99 

100<!-- x-release-please-end -->100<!-- x-release-please-end -->

899client.with_options(http_client=DefaultHttpxClient(...))899client.with_options(http_client=DefaultHttpxClient(...))

900```900```

901 901 

902#### Mutual TLS

903 

904Before configuring a client, review the

905[OpenAI Mutual TLS Beta Program](https://help.openai.com/en/articles/10876024-openai-mutual-tls-beta-program)

906for enrollment, currently supported endpoints, and certificate requirements.

907 

908For API-key authenticated HTTP requests that require mutual TLS (mTLS), configure

909a native [`ssl.SSLContext`](https://docs.python.org/3/library/ssl.html#ssl.SSLContext)

910and pass it through the custom HTTP client:

911 

912```python

913import os

914import ssl

915 

916from openai import OpenAI, DefaultHttpxClient

917 

918# Server trust is configured independently. Without `cafile`, this uses the

919# operating system's normal trusted certificate authorities.

920ssl_context = ssl.create_default_context(

921 cafile=os.environ.get("OPENAI_MTLS_CA_BUNDLE"),

922)

923ssl_context.load_cert_chain(

924 # This PEM must contain the leaf certificate first, followed by every

925 # intermediate certificate needed to reach the server's trust anchor.

926 certfile=os.environ["OPENAI_MTLS_CERTIFICATE_CHAIN"],

927 keyfile=os.environ["OPENAI_MTLS_PRIVATE_KEY"],

928 password=os.environ.get("OPENAI_MTLS_PRIVATE_KEY_PASSWORD"),

929)

930 

931client = OpenAI(

932 api_key=os.environ["OPENAI_API_KEY"],

933 # A custom HTTP client does not tell the SDK that mTLS is configured, so

934 # select the mTLS endpoint explicitly. Preserve an EU or custom override.

935 base_url=os.environ.get(

936 "OPENAI_BASE_URL",

937 "https://mtls.api.openai.com/v1",

938 ),

939 # A client certificate belongs to the HTTP client, not the base URL.

940 # Disable redirects so it cannot follow a response to another origin.

941 http_client=DefaultHttpxClient(

942 verify=ssl_context,

943 follow_redirects=False,

944 ),

945)

946```

947 

948The async configuration is equivalent:

949 

950```python

951import os

952import ssl

953 

954from openai import AsyncOpenAI, DefaultAsyncHttpxClient

955 

956ssl_context = ssl.create_default_context(

957 cafile=os.environ.get("OPENAI_MTLS_CA_BUNDLE"),

958)

959ssl_context.load_cert_chain(

960 certfile=os.environ["OPENAI_MTLS_CERTIFICATE_CHAIN"],

961 keyfile=os.environ["OPENAI_MTLS_PRIVATE_KEY"],

962 password=os.environ.get("OPENAI_MTLS_PRIVATE_KEY_PASSWORD"),

963)

964 

965client = AsyncOpenAI(

966 api_key=os.environ["OPENAI_API_KEY"],

967 base_url=os.environ.get(

968 "OPENAI_BASE_URL",

969 "https://mtls.api.openai.com/v1",

970 ),

971 http_client=DefaultAsyncHttpxClient(

972 verify=ssl_context,

973 follow_redirects=False,

974 ),

975)

976```

977 

978Experimental HTTPX2 uses the same native `SSLContext`. Install the optional

979extra with `pip install 'openai[httpx2]'`, then use `DefaultHttpx2Client` or

980`DefaultAsyncHttpx2Client` in place of the corresponding HTTPX client above:

981 

982```python

983from openai import OpenAI, DefaultHttpx2Client

984 

985client = OpenAI(

986 api_key=os.environ["OPENAI_API_KEY"],

987 base_url=os.environ.get(

988 "OPENAI_BASE_URL",

989 "https://mtls.api.openai.com/v1",

990 ),

991 http_client=DefaultHttpx2Client(

992 verify=ssl_context,

993 follow_redirects=False,

994 ),

995)

996```

997 

998See the complete [sync HTTPX2](examples/mtls_httpx2.py) and

999[async HTTPX2](examples/mtls_httpx2_async.py) examples.

1000 

1001The certificate-bearing HTTP client is transport-wide. Dedicate it to the

1002selected mTLS origin; do not reuse it for other services or pass it through

1003`with_options()` with a different `base_url`. If redirects are required, add an

1004HTTPX request hook that rejects requests whose scheme, host, or port differs

1005from the configured mTLS origin before enabling `follow_redirects`.

1006 

1007`SSLContext.load_cert_chain()` raises during setup for unreadable or malformed

1008files and for a private key that does not match the leaf certificate. Certificate

1009expiry, key usage, extended key usage, SAN, and trust policy remain TLS server

1010decisions. OpenAI does not fetch missing intermediates through AIA, so provide a

1011complete, leaf-first client-chain PEM. Intermediate-chain support is currently

1012enabled by request. Until it is enabled for your organization, use a client leaf

1013certificate directly signed by the uploaded CA.

1014 

1015For certificate rotation, build a new `SSLContext`, HTTP client, and `OpenAI` or

1016`AsyncOpenAI` client. This creates a fresh connection pool; close the old SDK

1017client after its in-flight requests finish. Do not assume existing TLS

1018connections will renegotiate.

1019 

1020This recipe applies to ordinary API-key HTTP traffic. It does not implement

1021certificate-only X.509 workload identity, token exchange, or Realtime WebSocket

1022mTLS.

1023 

902### Managing HTTP resources1024### Managing HTTP resources

903 1025 

904By default the library closes underlying HTTP connections whenever the client is [garbage collected](https://docs.python.org/3/reference/datamodel.html#object.__del__). You can manually close the client using the `.close()` method if desired, or with a context manager that closes when exiting.1026By default the library closes underlying HTTP connections whenever the client is [garbage collected](https://docs.python.org/3/reference/datamodel.html#object.__del__). You can manually close the client using the `.close()` method if desired, or with a context manager that closes when exiting.

9436 9436 

9437 - `"incomplete"`9437 - `"incomplete"`

9438 9438 

9439 - `FunctionCallOutput object { call_id, output, type, 4 more }`9439 - `FunctionCallOutput object { call_id, output, type, 6 more }`

9440 9440 

9441 The output of a function tool call.9441 The output of a function tool call.

9442 9442 

9608 9608 

9609 - `"program"`9609 - `"program"`

9610 9610 

9611 - `name: optional string`

9612 

9613 The name of the tool that produced the output.

9614 

9615 - `namespace: optional string`

9616 

9617 The namespace of the tool that produced the output.

9618 

9611 - `status: optional "in_progress" or "completed" or "incomplete"`9619 - `status: optional "in_progress" or "completed" or "incomplete"`

9612 9620 

9613 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.9621 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

13575 13583 

13576 - `"incomplete"`13584 - `"incomplete"`

13577 13585 

13578 - `FunctionCallOutput object { id, call_id, output, 5 more }`13586 - `FunctionCallOutput object { id, call_id, output, 7 more }`

13579 13587 

13580 - `id: string`13588 - `id: string`

13581 13589 

13663 13671 

13664 The identifier of the actor that created the item.13672 The identifier of the actor that created the item.

13665 13673 

13674 - `name: optional string`

13675 

13676 The name of the tool that produced the output.

13677 

13678 - `namespace: optional string`

13679 

13680 The namespace of the tool that produced the output.

13681 

13666 - `AgentMessage object { id, author, content, 3 more }`13682 - `AgentMessage object { id, author, content, 3 more }`

13667 13683 

13668 - `id: string`13684 - `id: string`

18875 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.18891 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

18876 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).18892 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).

18877 18893 

18878 - `service_tier: optional "auto" or "default" or "flex" or 2 more`18894 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

18879 18895 

18880 Specifies the processing type used for serving the request.18896 Specifies the processing type used for serving the request.

18881 18897 

18882 - 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'.18898 - 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'.

18883 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.18899 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

18884 - 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.18900 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

18901 - 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.

18885 - When not set, the default behavior is 'auto'.18902 - When not set, the default behavior is 'auto'.

18886 18903 

18887 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.18904 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.

18896 18913 

18897 - `"priority"`18914 - `"priority"`

18898 18915 

18916 - `"fast"`

18917 

18899 - `status: optional BetaResponseStatus`18918 - `status: optional BetaResponseStatus`

18900 18919 

18901 The status of the response generation. One of `completed`, `failed`,18920 The status of the response generation. One of `completed`, `failed`,

20559 20578 

20560 - `"incomplete"`20579 - `"incomplete"`

20561 20580 

20562 - `FunctionCallOutput object { call_id, output, type, 4 more }`20581 - `FunctionCallOutput object { call_id, output, type, 6 more }`

20563 20582 

20564 The output of a function tool call.20583 The output of a function tool call.

20565 20584 

20731 20750 

20732 - `"program"`20751 - `"program"`

20733 20752 

20753 - `name: optional string`

20754 

20755 The name of the tool that produced the output.

20756 

20757 - `namespace: optional string`

20758 

20759 The namespace of the tool that produced the output.

20760 

20734 - `status: optional "in_progress" or "completed" or "incomplete"`20761 - `status: optional "in_progress" or "completed" or "incomplete"`

20735 20762 

20736 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.20763 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

24355 24382 

24356 - `"24h"`24383 - `"24h"`

24357 24384 

24358- `service_tier: optional "auto" or "default" or "flex" or "priority"`24385- `service_tier: optional "auto" or "default" or "fast" or 2 more`

24359 24386 

24360 The service tier to use for this request.24387 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'.

24388 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.

24361 24389 

24362 - `"auto"`24390 - `"auto"`

24363 24391 

24364 - `"default"`24392 - `"default"`

24365 24393 

24394 - `"fast"`

24395 

24366 - `"flex"`24396 - `"flex"`

24367 24397 

24368 - `"priority"`24398 - `"priority"`

27455 27485 

27456 The canonical name of the agent that produced this item.27486 The canonical name of the agent that produced this item.

27457 27487 

27458 - `FunctionCallOutput object { call_id, output, type, 4 more }`27488 - `FunctionCallOutput object { call_id, output, type, 6 more }`

27459 27489 

27460 The output of a function tool call.27490 The output of a function tool call.

27461 27491 

27531 27561 

27532 - `"program"`27562 - `"program"`

27533 27563 

27564 - `name: optional string`

27565 

27566 The name of the tool that produced the output.

27567 

27568 - `namespace: optional string`

27569 

27570 The namespace of the tool that produced the output.

27571 

27534 - `status: optional "in_progress" or "completed" or "incomplete"`27572 - `status: optional "in_progress" or "completed" or "incomplete"`

27535 27573 

27536 The status of the item. One of `in_progress`, `completed`, or27574 The status of the item. One of `in_progress`, `completed`, or

30383 30421 

30384 - `"incomplete"`30422 - `"incomplete"`

30385 30423 

30386 - `FunctionCallOutput object { call_id, output, type, 4 more }`30424 - `FunctionCallOutput object { call_id, output, type, 6 more }`

30387 30425 

30388 The output of a function tool call.30426 The output of a function tool call.

30389 30427 

30555 30593 

30556 - `"program"`30594 - `"program"`

30557 30595 

30596 - `name: optional string`

30597 

30598 The name of the tool that produced the output.

30599 

30600 - `namespace: optional string`

30601 

30602 The namespace of the tool that produced the output.

30603 

30558 - `status: optional "in_progress" or "completed" or "incomplete"`30604 - `status: optional "in_progress" or "completed" or "incomplete"`

30559 30605 

30560 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.30606 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

34598 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.34644 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

34599 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).34645 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).

34600 34646 

34601- `service_tier: optional "auto" or "default" or "flex" or 2 more`34647- `service_tier: optional "auto" or "default" or "flex" or 3 more`

34602 34648 

34603 Specifies the processing type used for serving the request.34649 Specifies the processing type used for serving the request.

34604 34650 

34605 - 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'.34651 - 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'.

34606 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.34652 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

34607 - 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.34653 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

34654 - 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.

34608 - When not set, the default behavior is 'auto'.34655 - When not set, the default behavior is 'auto'.

34609 34656 

34610 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.34657 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.

34619 34666 

34620 - `"priority"`34667 - `"priority"`

34621 34668 

34669 - `"fast"`

34670 

34622- `store: optional boolean`34671- `store: optional boolean`

34623 34672 

34624 Whether to store the generated model response for later retrieval via34673 Whether to store the generated model response for later retrieval via

37068 37117 

37069 - `"incomplete"`37118 - `"incomplete"`

37070 37119 

37071 - `FunctionCallOutput object { call_id, output, type, 4 more }`37120 - `FunctionCallOutput object { call_id, output, type, 6 more }`

37072 37121 

37073 The output of a function tool call.37122 The output of a function tool call.

37074 37123 

37240 37289 

37241 - `"program"`37290 - `"program"`

37242 37291 

37292 - `name: optional string`

37293 

37294 The name of the tool that produced the output.

37295 

37296 - `namespace: optional string`

37297 

37298 The namespace of the tool that produced the output.

37299 

37243 - `status: optional "in_progress" or "completed" or "incomplete"`37300 - `status: optional "in_progress" or "completed" or "incomplete"`

37244 37301 

37245 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.37302 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

41207 41264 

41208 - `"incomplete"`41265 - `"incomplete"`

41209 41266 

41210 - `FunctionCallOutput object { id, call_id, output, 5 more }`41267 - `FunctionCallOutput object { id, call_id, output, 7 more }`

41211 41268 

41212 - `id: string`41269 - `id: string`

41213 41270 

41295 41352 

41296 The identifier of the actor that created the item.41353 The identifier of the actor that created the item.

41297 41354 

41355 - `name: optional string`

41356 

41357 The name of the tool that produced the output.

41358 

41359 - `namespace: optional string`

41360 

41361 The namespace of the tool that produced the output.

41362 

41298 - `AgentMessage object { id, author, content, 3 more }`41363 - `AgentMessage object { id, author, content, 3 more }`

41299 41364 

41300 - `id: string`41365 - `id: string`

46507 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.46572 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

46508 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).46573 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).

46509 46574 

46510 - `service_tier: optional "auto" or "default" or "flex" or 2 more`46575 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

46511 46576 

46512 Specifies the processing type used for serving the request.46577 Specifies the processing type used for serving the request.

46513 46578 

46514 - 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'.46579 - 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'.

46515 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.46580 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

46516 - 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.46581 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

46582 - 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.

46517 - When not set, the default behavior is 'auto'.46583 - When not set, the default behavior is 'auto'.

46518 46584 

46519 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.46585 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.

46528 46594 

46529 - `"priority"`46595 - `"priority"`

46530 46596 

46597 - `"fast"`

46598 

46531 - `status: optional BetaResponseStatus`46599 - `status: optional BetaResponseStatus`

46532 46600 

46533 The status of the response generation. One of `completed`, `failed`,46601 The status of the response generation. One of `completed`, `failed`,

48868 48936 

48869 - `"incomplete"`48937 - `"incomplete"`

48870 48938 

48871 - `FunctionCallOutput object { call_id, output, type, 4 more }`48939 - `FunctionCallOutput object { call_id, output, type, 6 more }`

48872 48940 

48873 The output of a function tool call.48941 The output of a function tool call.

48874 48942 

49040 49108 

49041 - `"program"`49109 - `"program"`

49042 49110 

49111 - `name: optional string`

49112 

49113 The name of the tool that produced the output.

49114 

49115 - `namespace: optional string`

49116 

49117 The namespace of the tool that produced the output.

49118 

49043 - `status: optional "in_progress" or "completed" or "incomplete"`49119 - `status: optional "in_progress" or "completed" or "incomplete"`

49044 49120 

49045 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.49121 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

53007 53083 

53008 - `"incomplete"`53084 - `"incomplete"`

53009 53085 

53010 - `FunctionCallOutput object { id, call_id, output, 5 more }`53086 - `FunctionCallOutput object { id, call_id, output, 7 more }`

53011 53087 

53012 - `id: string`53088 - `id: string`

53013 53089 

53095 53171 

53096 The identifier of the actor that created the item.53172 The identifier of the actor that created the item.

53097 53173 

53174 - `name: optional string`

53175 

53176 The name of the tool that produced the output.

53177 

53178 - `namespace: optional string`

53179 

53180 The namespace of the tool that produced the output.

53181 

53098 - `AgentMessage object { id, author, content, 3 more }`53182 - `AgentMessage object { id, author, content, 3 more }`

53099 53183 

53100 - `id: string`53184 - `id: string`

58307 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.58391 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

58308 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).58392 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).

58309 58393 

58310 - `service_tier: optional "auto" or "default" or "flex" or 2 more`58394 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

58311 58395 

58312 Specifies the processing type used for serving the request.58396 Specifies the processing type used for serving the request.

58313 58397 

58314 - 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'.58398 - 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'.

58315 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.58399 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

58316 - 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.58400 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

58401 - 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.

58317 - When not set, the default behavior is 'auto'.58402 - When not set, the default behavior is 'auto'.

58318 58403 

58319 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.58404 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.

58328 58413 

58329 - `"priority"`58414 - `"priority"`

58330 58415 

58416 - `"fast"`

58417 

58331 - `status: optional BetaResponseStatus`58418 - `status: optional BetaResponseStatus`

58332 58419 

58333 The status of the response generation. One of `completed`, `failed`,58420 The status of the response generation. One of `completed`, `failed`,

61847 61934 

61848 The canonical name of the agent that produced this item.61935 The canonical name of the agent that produced this item.

61849 61936 

61850 - `FunctionCallOutput object { call_id, output, type, 4 more }`61937 - `FunctionCallOutput object { call_id, output, type, 6 more }`

61851 61938 

61852 The output of a function tool call.61939 The output of a function tool call.

61853 61940 

61923 62010 

61924 - `"program"`62011 - `"program"`

61925 62012 

62013 - `name: optional string`

62014 

62015 The name of the tool that produced the output.

62016 

62017 - `namespace: optional string`

62018 

62019 The namespace of the tool that produced the output.

62020 

61926 - `status: optional "in_progress" or "completed" or "incomplete"`62021 - `status: optional "in_progress" or "completed" or "incomplete"`

61927 62022 

61928 The status of the item. One of `in_progress`, `completed`, or62023 The status of the item. One of `in_progress`, `completed`, or

65490 65585 

65491 - `"incomplete"`65586 - `"incomplete"`

65492 65587 

65493 - `FunctionCallOutput object { call_id, output, type, 4 more }`65588 - `FunctionCallOutput object { call_id, output, type, 6 more }`

65494 65589 

65495 The output of a function tool call.65590 The output of a function tool call.

65496 65591 

65662 65757 

65663 - `"program"`65758 - `"program"`

65664 65759 

65760 - `name: optional string`

65761 

65762 The name of the tool that produced the output.

65763 

65764 - `namespace: optional string`

65765 

65766 The namespace of the tool that produced the output.

65767 

65665 - `status: optional "in_progress" or "completed" or "incomplete"`65768 - `status: optional "in_progress" or "completed" or "incomplete"`

65666 65769 

65667 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.65770 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

69629 69732 

69630 - `"incomplete"`69733 - `"incomplete"`

69631 69734 

69632 - `FunctionCallOutput object { id, call_id, output, 5 more }`69735 - `FunctionCallOutput object { id, call_id, output, 7 more }`

69633 69736 

69634 - `id: string`69737 - `id: string`

69635 69738 

69717 69820 

69718 The identifier of the actor that created the item.69821 The identifier of the actor that created the item.

69719 69822 

69823 - `name: optional string`

69824 

69825 The name of the tool that produced the output.

69826 

69827 - `namespace: optional string`

69828 

69829 The namespace of the tool that produced the output.

69830 

69720 - `AgentMessage object { id, author, content, 3 more }`69831 - `AgentMessage object { id, author, content, 3 more }`

69721 69832 

69722 - `id: string`69833 - `id: string`

74929 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.75040 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

74930 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).75041 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).

74931 75042 

74932 - `service_tier: optional "auto" or "default" or "flex" or 2 more`75043 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

74933 75044 

74934 Specifies the processing type used for serving the request.75045 Specifies the processing type used for serving the request.

74935 75046 

74936 - 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'.75047 - 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'.

74937 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.75048 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

74938 - 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.75049 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

75050 - 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.

74939 - When not set, the default behavior is 'auto'.75051 - When not set, the default behavior is 'auto'.

74940 75052 

74941 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.75053 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.

74950 75062 

74951 - `"priority"`75063 - `"priority"`

74952 75064 

75065 - `"fast"`

75066 

74953 - `status: optional BetaResponseStatus`75067 - `status: optional BetaResponseStatus`

74954 75068 

74955 The status of the response generation. One of `completed`, `failed`,75069 The status of the response generation. One of `completed`, `failed`,

76513 76627 

76514 - `"incomplete"`76628 - `"incomplete"`

76515 76629 

76516 - `FunctionCallOutput object { call_id, output, type, 4 more }`76630 - `FunctionCallOutput object { call_id, output, type, 6 more }`

76517 76631 

76518 The output of a function tool call.76632 The output of a function tool call.

76519 76633 

76685 76799 

76686 - `"program"`76800 - `"program"`

76687 76801 

76802 - `name: optional string`

76803 

76804 The name of the tool that produced the output.

76805 

76806 - `namespace: optional string`

76807 

76808 The namespace of the tool that produced the output.

76809 

76688 - `status: optional "in_progress" or "completed" or "incomplete"`76810 - `status: optional "in_progress" or "completed" or "incomplete"`

76689 76811 

76690 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.76812 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

80652 80774 

80653 - `"incomplete"`80775 - `"incomplete"`

80654 80776 

80655 - `FunctionCallOutput object { id, call_id, output, 5 more }`80777 - `FunctionCallOutput object { id, call_id, output, 7 more }`

80656 80778 

80657 - `id: string`80779 - `id: string`

80658 80780 

80740 80862 

80741 The identifier of the actor that created the item.80863 The identifier of the actor that created the item.

80742 80864 

80865 - `name: optional string`

80866 

80867 The name of the tool that produced the output.

80868 

80869 - `namespace: optional string`

80870 

80871 The namespace of the tool that produced the output.

80872 

80743 - `AgentMessage object { id, author, content, 3 more }`80873 - `AgentMessage object { id, author, content, 3 more }`

80744 80874 

80745 - `id: string`80875 - `id: string`

85952 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.86082 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

85953 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).86083 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).

85954 86084 

85955 - `service_tier: optional "auto" or "default" or "flex" or 2 more`86085 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

85956 86086 

85957 Specifies the processing type used for serving the request.86087 Specifies the processing type used for serving the request.

85958 86088 

85959 - 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'.86089 - 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'.

85960 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.86090 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

85961 - 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.86091 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

86092 - 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.

85962 - When not set, the default behavior is 'auto'.86093 - When not set, the default behavior is 'auto'.

85963 86094 

85964 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.86095 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.

85973 86104 

85974 - `"priority"`86105 - `"priority"`

85975 86106 

86107 - `"fast"`

86108 

85976 - `status: optional BetaResponseStatus`86109 - `status: optional BetaResponseStatus`

85977 86110 

85978 The status of the response generation. One of `completed`, `failed`,86111 The status of the response generation. One of `completed`, `failed`,

87997 88130 

87998 - `"incomplete"`88131 - `"incomplete"`

87999 88132 

88000 - `FunctionCallOutput object { call_id, output, type, 4 more }`88133 - `FunctionCallOutput object { call_id, output, type, 6 more }`

88001 88134 

88002 The output of a function tool call.88135 The output of a function tool call.

88003 88136 

88169 88302 

88170 - `"program"`88303 - `"program"`

88171 88304 

88305 - `name: optional string`

88306 

88307 The name of the tool that produced the output.

88308 

88309 - `namespace: optional string`

88310 

88311 The namespace of the tool that produced the output.

88312 

88172 - `status: optional "in_progress" or "completed" or "incomplete"`88313 - `status: optional "in_progress" or "completed" or "incomplete"`

88173 88314 

88174 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.88315 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

92136 92277 

92137 - `"incomplete"`92278 - `"incomplete"`

92138 92279 

92139 - `FunctionCallOutput object { id, call_id, output, 5 more }`92280 - `FunctionCallOutput object { id, call_id, output, 7 more }`

92140 92281 

92141 - `id: string`92282 - `id: string`

92142 92283 

92224 92365 

92225 The identifier of the actor that created the item.92366 The identifier of the actor that created the item.

92226 92367 

92368 - `name: optional string`

92369 

92370 The name of the tool that produced the output.

92371 

92372 - `namespace: optional string`

92373 

92374 The namespace of the tool that produced the output.

92375 

92227 - `AgentMessage object { id, author, content, 3 more }`92376 - `AgentMessage object { id, author, content, 3 more }`

92228 92377 

92229 - `id: string`92378 - `id: string`

97436 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.97585 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

97437 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).97586 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).

97438 97587 

97439 - `service_tier: optional "auto" or "default" or "flex" or 2 more`97588 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

97440 97589 

97441 Specifies the processing type used for serving the request.97590 Specifies the processing type used for serving the request.

97442 97591 

97443 - 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'.97592 - 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'.

97444 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.97593 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

97445 - 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.97594 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

97595 - 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.

97446 - When not set, the default behavior is 'auto'.97596 - When not set, the default behavior is 'auto'.

97447 97597 

97448 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.97598 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.

97457 97607 

97458 - `"priority"`97608 - `"priority"`

97459 97609 

97610 - `"fast"`

97611 

97460 - `status: optional BetaResponseStatus`97612 - `status: optional BetaResponseStatus`

97461 97613 

97462 The status of the response generation. One of `completed`, `failed`,97614 The status of the response generation. One of `completed`, `failed`,

98928 99080 

98929 - `"incomplete"`99081 - `"incomplete"`

98930 99082 

98931 - `FunctionCallOutput object { call_id, output, type, 4 more }`99083 - `FunctionCallOutput object { call_id, output, type, 6 more }`

98932 99084 

98933 The output of a function tool call.99085 The output of a function tool call.

98934 99086 

99100 99252 

99101 - `"program"`99253 - `"program"`

99102 99254 

99255 - `name: optional string`

99256 

99257 The name of the tool that produced the output.

99258 

99259 - `namespace: optional string`

99260 

99261 The namespace of the tool that produced the output.

99262 

99103 - `status: optional "in_progress" or "completed" or "incomplete"`99263 - `status: optional "in_progress" or "completed" or "incomplete"`

99104 99264 

99105 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.99265 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

103067 103227 

103068 - `"incomplete"`103228 - `"incomplete"`

103069 103229 

103070 - `FunctionCallOutput object { id, call_id, output, 5 more }`103230 - `FunctionCallOutput object { id, call_id, output, 7 more }`

103071 103231 

103072 - `id: string`103232 - `id: string`

103073 103233 

103155 103315 

103156 The identifier of the actor that created the item.103316 The identifier of the actor that created the item.

103157 103317 

103318 - `name: optional string`

103319 

103320 The name of the tool that produced the output.

103321 

103322 - `namespace: optional string`

103323 

103324 The namespace of the tool that produced the output.

103325 

103158 - `AgentMessage object { id, author, content, 3 more }`103326 - `AgentMessage object { id, author, content, 3 more }`

103159 103327 

103160 - `id: string`103328 - `id: string`

108367 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.108535 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

108368 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).108536 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).

108369 108537 

108370 - `service_tier: optional "auto" or "default" or "flex" or 2 more`108538 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

108371 108539 

108372 Specifies the processing type used for serving the request.108540 Specifies the processing type used for serving the request.

108373 108541 

108374 - 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'.108542 - 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'.

108375 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.108543 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

108376 - 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.108544 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

108545 - 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.

108377 - When not set, the default behavior is 'auto'.108546 - When not set, the default behavior is 'auto'.

108378 108547 

108379 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.108548 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.

108388 108557 

108389 - `"priority"`108558 - `"priority"`

108390 108559 

108560 - `"fast"`

108561 

108391 - `status: optional BetaResponseStatus`108562 - `status: optional BetaResponseStatus`

108392 108563 

108393 The status of the response generation. One of `completed`, `failed`,108564 The status of the response generation. One of `completed`, `failed`,

110156 110327 

110157 - `"incomplete"`110328 - `"incomplete"`

110158 110329 

110159 - `FunctionCallOutput object { call_id, output, type, 4 more }`110330 - `FunctionCallOutput object { call_id, output, type, 6 more }`

110160 110331 

110161 The output of a function tool call.110332 The output of a function tool call.

110162 110333 

110328 110499 

110329 - `"program"`110500 - `"program"`

110330 110501 

110502 - `name: optional string`

110503 

110504 The name of the tool that produced the output.

110505 

110506 - `namespace: optional string`

110507 

110508 The namespace of the tool that produced the output.

110509 

110331 - `status: optional "in_progress" or "completed" or "incomplete"`110510 - `status: optional "in_progress" or "completed" or "incomplete"`

110332 110511 

110333 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.110512 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

114295 114474 

114296 - `"incomplete"`114475 - `"incomplete"`

114297 114476 

114298 - `FunctionCallOutput object { id, call_id, output, 5 more }`114477 - `FunctionCallOutput object { id, call_id, output, 7 more }`

114299 114478 

114300 - `id: string`114479 - `id: string`

114301 114480 

114383 114562 

114384 The identifier of the actor that created the item.114563 The identifier of the actor that created the item.

114385 114564 

114565 - `name: optional string`

114566 

114567 The name of the tool that produced the output.

114568 

114569 - `namespace: optional string`

114570 

114571 The namespace of the tool that produced the output.

114572 

114386 - `AgentMessage object { id, author, content, 3 more }`114573 - `AgentMessage object { id, author, content, 3 more }`

114387 114574 

114388 - `id: string`114575 - `id: string`

119595 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.119782 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

119596 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).119783 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).

119597 119784 

119598 - `service_tier: optional "auto" or "default" or "flex" or 2 more`119785 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

119599 119786 

119600 Specifies the processing type used for serving the request.119787 Specifies the processing type used for serving the request.

119601 119788 

119602 - 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'.119789 - 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'.

119603 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.119790 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

119604 - 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.119791 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

119792 - 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.

119605 - When not set, the default behavior is 'auto'.119793 - When not set, the default behavior is 'auto'.

119606 119794 

119607 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.119795 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.

119616 119804 

119617 - `"priority"`119805 - `"priority"`

119618 119806 

119807 - `"fast"`

119808 

119619 - `status: optional BetaResponseStatus`119809 - `status: optional BetaResponseStatus`

119620 119810 

119621 The status of the response generation. One of `completed`, `failed`,119811 The status of the response generation. One of `completed`, `failed`,

120956 121146 

120957 - `"incomplete"`121147 - `"incomplete"`

120958 121148 

120959 - `FunctionCallOutput object { call_id, output, type, 4 more }`121149 - `FunctionCallOutput object { call_id, output, type, 6 more }`

120960 121150 

120961 The output of a function tool call.121151 The output of a function tool call.

120962 121152 

121128 121318 

121129 - `"program"`121319 - `"program"`

121130 121320 

121321 - `name: optional string`

121322 

121323 The name of the tool that produced the output.

121324 

121325 - `namespace: optional string`

121326 

121327 The namespace of the tool that produced the output.

121328 

121131 - `status: optional "in_progress" or "completed" or "incomplete"`121329 - `status: optional "in_progress" or "completed" or "incomplete"`

121132 121330 

121133 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.121331 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

125095 125293 

125096 - `"incomplete"`125294 - `"incomplete"`

125097 125295 

125098 - `FunctionCallOutput object { id, call_id, output, 5 more }`125296 - `FunctionCallOutput object { id, call_id, output, 7 more }`

125099 125297 

125100 - `id: string`125298 - `id: string`

125101 125299 

125183 125381 

125184 The identifier of the actor that created the item.125382 The identifier of the actor that created the item.

125185 125383 

125384 - `name: optional string`

125385 

125386 The name of the tool that produced the output.

125387 

125388 - `namespace: optional string`

125389 

125390 The namespace of the tool that produced the output.

125391 

125186 - `AgentMessage object { id, author, content, 3 more }`125392 - `AgentMessage object { id, author, content, 3 more }`

125187 125393 

125188 - `id: string`125394 - `id: string`

130395 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.130601 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

130396 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).130602 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).

130397 130603 

130398 - `service_tier: optional "auto" or "default" or "flex" or 2 more`130604 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

130399 130605 

130400 Specifies the processing type used for serving the request.130606 Specifies the processing type used for serving the request.

130401 130607 

130402 - 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'.130608 - 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'.

130403 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.130609 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

130404 - 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.130610 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

130611 - 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.

130405 - When not set, the default behavior is 'auto'.130612 - When not set, the default behavior is 'auto'.

130406 130613 

130407 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.130614 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.

130416 130623 

130417 - `"priority"`130624 - `"priority"`

130418 130625 

130626 - `"fast"`

130627 

130419 - `status: optional BetaResponseStatus`130628 - `status: optional BetaResponseStatus`

130420 130629 

130421 The status of the response generation. One of `completed`, `failed`,130630 The status of the response generation. One of `completed`, `failed`,

131663 131872 

131664 - `"incomplete"`131873 - `"incomplete"`

131665 131874 

131666 - `FunctionCallOutput object { call_id, output, type, 4 more }`131875 - `FunctionCallOutput object { call_id, output, type, 6 more }`

131667 131876 

131668 The output of a function tool call.131877 The output of a function tool call.

131669 131878 

131835 132044 

131836 - `"program"`132045 - `"program"`

131837 132046 

132047 - `name: optional string`

132048 

132049 The name of the tool that produced the output.

132050 

132051 - `namespace: optional string`

132052 

132053 The namespace of the tool that produced the output.

132054 

131838 - `status: optional "in_progress" or "completed" or "incomplete"`132055 - `status: optional "in_progress" or "completed" or "incomplete"`

131839 132056 

131840 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.132057 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

136473 136690 

136474 - `"incomplete"`136691 - `"incomplete"`

136475 136692 

136476 - `FunctionCallOutput object { call_id, output, type, 4 more }`136693 - `FunctionCallOutput object { call_id, output, type, 6 more }`

136477 136694 

136478 The output of a function tool call.136695 The output of a function tool call.

136479 136696 

136645 136862 

136646 - `"program"`136863 - `"program"`

136647 136864 

136865 - `name: optional string`

136866 

136867 The name of the tool that produced the output.

136868 

136869 - `namespace: optional string`

136870 

136871 The namespace of the tool that produced the output.

136872 

136648 - `status: optional "in_progress" or "completed" or "incomplete"`136873 - `status: optional "in_progress" or "completed" or "incomplete"`

136649 136874 

136650 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.136875 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

141544 141769 

141545 - `"incomplete"`141770 - `"incomplete"`

141546 141771 

141547 - `FunctionCallOutput object { id, call_id, output, 5 more }`141772 - `FunctionCallOutput object { id, call_id, output, 7 more }`

141548 141773 

141549 - `id: string`141774 - `id: string`

141550 141775 

141730 141955 

141731 The identifier of the actor that created the item.141956 The identifier of the actor that created the item.

141732 141957 

141958 - `name: optional string`

141959 

141960 The name of the tool that produced the output.

141961 

141962 - `namespace: optional string`

141963 

141964 The namespace of the tool that produced the output.

141965 

141733 - `AgentMessage object { id, author, content, 3 more }`141966 - `AgentMessage object { id, author, content, 3 more }`

141734 141967 

141735 - `id: string`141968 - `id: string`

146164 146397 

146165 - `"incomplete"`146398 - `"incomplete"`

146166 146399 

146167 - `FunctionCallOutput object { id, call_id, output, 5 more }`146400 - `FunctionCallOutput object { id, call_id, output, 7 more }`

146168 146401 

146169 - `id: string`146402 - `id: string`

146170 146403 

146350 146583 

146351 The identifier of the actor that created the item.146584 The identifier of the actor that created the item.

146352 146585 

146586 - `name: optional string`

146587 

146588 The name of the tool that produced the output.

146589 

146590 - `namespace: optional string`

146591 

146592 The namespace of the tool that produced the output.

146593 

146353 - `AgentMessage object { id, author, content, 3 more }`146594 - `AgentMessage object { id, author, content, 3 more }`

146354 146595 

146355 - `id: string`146596 - `id: string`

150806 151047 

150807 - `"incomplete"`151048 - `"incomplete"`

150808 151049 

150809 - `FunctionCallOutput object { id, call_id, output, 5 more }`151050 - `FunctionCallOutput object { id, call_id, output, 7 more }`

150810 151051 

150811 - `id: string`151052 - `id: string`

150812 151053 

150992 151233 

150993 The identifier of the actor that created the item.151234 The identifier of the actor that created the item.

150994 151235 

151236 - `name: optional string`

151237 

151238 The name of the tool that produced the output.

151239 

151240 - `namespace: optional string`

151241 

151242 The namespace of the tool that produced the output.

151243 

150995 - `AgentMessage object { id, author, content, 3 more }`151244 - `AgentMessage object { id, author, content, 3 more }`

150996 151245 

150997 - `id: string`151246 - `id: string`

156734 156983 

156735 - `"incomplete"`156984 - `"incomplete"`

156736 156985 

156737 - `FunctionCallOutput object { call_id, output, type, 4 more }`156986 - `FunctionCallOutput object { call_id, output, type, 6 more }`

156738 156987 

156739 The output of a function tool call.156988 The output of a function tool call.

156740 156989 

156906 157155 

156907 - `"program"`157156 - `"program"`

156908 157157 

157158 - `name: optional string`

157159 

157160 The name of the tool that produced the output.

157161 

157162 - `namespace: optional string`

157163 

157164 The namespace of the tool that produced the output.

157165 

156909 - `status: optional "in_progress" or "completed" or "incomplete"`157166 - `status: optional "in_progress" or "completed" or "incomplete"`

156910 157167 

156911 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.157168 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

160873 161130 

160874 - `"incomplete"`161131 - `"incomplete"`

160875 161132 

160876 - `FunctionCallOutput object { id, call_id, output, 5 more }`161133 - `FunctionCallOutput object { id, call_id, output, 7 more }`

160877 161134 

160878 - `id: string`161135 - `id: string`

160879 161136 

160961 161218 

160962 The identifier of the actor that created the item.161219 The identifier of the actor that created the item.

160963 161220 

161221 - `name: optional string`

161222 

161223 The name of the tool that produced the output.

161224 

161225 - `namespace: optional string`

161226 

161227 The namespace of the tool that produced the output.

161228 

160964 - `AgentMessage object { id, author, content, 3 more }`161229 - `AgentMessage object { id, author, content, 3 more }`

160965 161230 

160966 - `id: string`161231 - `id: string`

166173 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.166438 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

166174 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).166439 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).

166175 166440 

166176 - `service_tier: optional "auto" or "default" or "flex" or 2 more`166441 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

166177 166442 

166178 Specifies the processing type used for serving the request.166443 Specifies the processing type used for serving the request.

166179 166444 

166180 - 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'.166445 - 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'.

166181 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.166446 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

166182 - 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.166447 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

166448 - 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.

166183 - When not set, the default behavior is 'auto'.166449 - When not set, the default behavior is 'auto'.

166184 166450 

166185 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.166451 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.

166194 166460 

166195 - `"priority"`166461 - `"priority"`

166196 166462 

166463 - `"fast"`

166464 

166197 - `status: optional BetaResponseStatus`166465 - `status: optional BetaResponseStatus`

166198 166466 

166199 The status of the response generation. One of `completed`, `failed`,166467 The status of the response generation. One of `completed`, `failed`,

168127 168395 

168128 - `"incomplete"`168396 - `"incomplete"`

168129 168397 

168130 - `FunctionCallOutput object { call_id, output, type, 4 more }`168398 - `FunctionCallOutput object { call_id, output, type, 6 more }`

168131 168399 

168132 The output of a function tool call.168400 The output of a function tool call.

168133 168401 

168299 168567 

168300 - `"program"`168568 - `"program"`

168301 168569 

168570 - `name: optional string`

168571 

168572 The name of the tool that produced the output.

168573 

168574 - `namespace: optional string`

168575 

168576 The namespace of the tool that produced the output.

168577 

168302 - `status: optional "in_progress" or "completed" or "incomplete"`168578 - `status: optional "in_progress" or "completed" or "incomplete"`

168303 168579 

168304 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.168580 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

172266 172542 

172267 - `"incomplete"`172543 - `"incomplete"`

172268 172544 

172269 - `FunctionCallOutput object { id, call_id, output, 5 more }`172545 - `FunctionCallOutput object { id, call_id, output, 7 more }`

172270 172546 

172271 - `id: string`172547 - `id: string`

172272 172548 

172354 172630 

172355 The identifier of the actor that created the item.172631 The identifier of the actor that created the item.

172356 172632 

172633 - `name: optional string`

172634 

172635 The name of the tool that produced the output.

172636 

172637 - `namespace: optional string`

172638 

172639 The namespace of the tool that produced the output.

172640 

172357 - `AgentMessage object { id, author, content, 3 more }`172641 - `AgentMessage object { id, author, content, 3 more }`

172358 172642 

172359 - `id: string`172643 - `id: string`

177566 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.177850 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

177567 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).177851 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).

177568 177852 

177569 - `service_tier: optional "auto" or "default" or "flex" or 2 more`177853 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

177570 177854 

177571 Specifies the processing type used for serving the request.177855 Specifies the processing type used for serving the request.

177572 177856 

177573 - 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'.177857 - 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'.

177574 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.177858 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

177575 - 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.177859 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

177860 - 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.

177576 - When not set, the default behavior is 'auto'.177861 - When not set, the default behavior is 'auto'.

177577 177862 

177578 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.177863 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.

177587 177872 

177588 - `"priority"`177873 - `"priority"`

177589 177874 

177875 - `"fast"`

177876 

177590 - `status: optional BetaResponseStatus`177877 - `status: optional BetaResponseStatus`

177591 177878 

177592 The status of the response generation. One of `completed`, `failed`,177879 The status of the response generation. One of `completed`, `failed`,

178221 A tool call to run a function. See the178508 A tool call to run a function. See the

178222 [function calling guide](/docs/guides/function-calling) for more information.178509 [function calling guide](/docs/guides/function-calling) for more information.

178223 178510 

178224 - `FunctionCallOutput object { id, call_id, output, 5 more }`178511 - `FunctionCallOutput object { id, call_id, output, 7 more }`

178225 178512 

178226 - `AgentMessage object { id, author, content, 3 more }`178513 - `AgentMessage object { id, author, content, 3 more }`

178227 178514 

180906 181193 

180907 - `"incomplete"`181194 - `"incomplete"`

180908 181195 

180909 - `FunctionCallOutput object { call_id, output, type, 4 more }`181196 - `FunctionCallOutput object { call_id, output, type, 6 more }`

180910 181197 

180911 The output of a function tool call.181198 The output of a function tool call.

180912 181199 

181078 181365 

181079 - `"program"`181366 - `"program"`

181080 181367 

181368 - `name: optional string`

181369 

181370 The name of the tool that produced the output.

181371 

181372 - `namespace: optional string`

181373 

181374 The namespace of the tool that produced the output.

181375 

181081 - `status: optional "in_progress" or "completed" or "incomplete"`181376 - `status: optional "in_progress" or "completed" or "incomplete"`

181082 181377 

181083 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.181378 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

185121 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.185416 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

185122 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).185417 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).

185123 185418 

185124 - `service_tier: optional "auto" or "default" or "flex" or 2 more`185419 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

185125 185420 

185126 Specifies the processing type used for serving the request.185421 Specifies the processing type used for serving the request.

185127 185422 

185128 - 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'.185423 - 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'.

185129 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.185424 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

185130 - 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.185425 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

185426 - 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.

185131 - When not set, the default behavior is 'auto'.185427 - When not set, the default behavior is 'auto'.

185132 185428 

185133 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.185429 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.

185142 185438 

185143 - `"priority"`185439 - `"priority"`

185144 185440 

185441 - `"fast"`

185442 

185145 - `store: optional boolean`185443 - `store: optional boolean`

185146 185444 

185147 Whether to store the generated model response for later retrieval via185445 Whether to store the generated model response for later retrieval via

186919 187217 

186920 - `"incomplete"`187218 - `"incomplete"`

186921 187219 

186922 - `FunctionCallOutput object { call_id, output, type, 4 more }`187220 - `FunctionCallOutput object { call_id, output, type, 6 more }`

186923 187221 

186924 The output of a function tool call.187222 The output of a function tool call.

186925 187223 

186993 187291 

186994 - `"program"`187292 - `"program"`

186995 187293 

187294 - `name: optional string`

187295 

187296 The name of the tool that produced the output.

187297 

187298 - `namespace: optional string`

187299 

187300 The namespace of the tool that produced the output.

187301 

186996 - `status: optional "in_progress" or "completed" or "incomplete"`187302 - `status: optional "in_progress" or "completed" or "incomplete"`

186997 187303 

186998 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.187304 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

191787 192093 

191788 - `"incomplete"`192094 - `"incomplete"`

191789 192095 

191790 - `FunctionCallOutput object { call_id, output, type, 4 more }`192096 - `FunctionCallOutput object { call_id, output, type, 6 more }`

191791 192097 

191792 The output of a function tool call.192098 The output of a function tool call.

191793 192099 

191959 192265 

191960 - `"program"`192266 - `"program"`

191961 192267 

192268 - `name: optional string`

192269 

192270 The name of the tool that produced the output.

192271 

192272 - `namespace: optional string`

192273 

192274 The namespace of the tool that produced the output.

192275 

191962 - `status: optional "in_progress" or "completed" or "incomplete"`192276 - `status: optional "in_progress" or "completed" or "incomplete"`

191963 192277 

191964 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.192278 The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.

195926 196240 

195927 - `"incomplete"`196241 - `"incomplete"`

195928 196242 

195929 - `FunctionCallOutput object { id, call_id, output, 5 more }`196243 - `FunctionCallOutput object { id, call_id, output, 7 more }`

195930 196244 

195931 - `id: string`196245 - `id: string`

195932 196246 

196014 196328 

196015 The identifier of the actor that created the item.196329 The identifier of the actor that created the item.

196016 196330 

196331 - `name: optional string`

196332 

196333 The name of the tool that produced the output.

196334 

196335 - `namespace: optional string`

196336 

196337 The namespace of the tool that produced the output.

196338 

196017 - `AgentMessage object { id, author, content, 3 more }`196339 - `AgentMessage object { id, author, content, 3 more }`

196018 196340 

196019 - `id: string`196341 - `id: string`

201226 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.201548 A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies.

201227 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).201549 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).

201228 201550 

201229 - `service_tier: optional "auto" or "default" or "flex" or 2 more`201551 - `service_tier: optional "auto" or "default" or "flex" or 3 more`

201230 201552 

201231 Specifies the processing type used for serving the request.201553 Specifies the processing type used for serving the request.

201232 201554 

201233 - 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'.201555 - 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'.

201234 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.201556 - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model.

201235 - 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.201557 - If set to '[flex](/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier.

201558 - 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.

201236 - When not set, the default behavior is 'auto'.201559 - When not set, the default behavior is 'auto'.

201237 201560 

201238 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.201561 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.

201247 201570 

201248 - `"priority"`201571 - `"priority"`

201249 201572 

201573 - `"fast"`

201574 

201250 - `status: optional BetaResponseStatus`201575 - `status: optional BetaResponseStatus`