openai 6.36.0 → 6.37.0

package/src/resources/images.ts

@@ -400,7 +400,15 @@ export interface ImageGenPartialImageEvent {
  */
 export type ImageGenStreamEvent = ImageGenPartialImageEvent | ImageGenCompletedEvent;
 
-export type ImageModel = 'gpt-image-1.5' | 'dall-e-2' | 'dall-e-3' | 'gpt-image-1' | 'gpt-image-1-mini';
+export type ImageModel =
+  | 'gpt-image-1'
+  | 'gpt-image-1-mini'
+  | 'gpt-image-2'
+  | 'gpt-image-2-2026-04-21'
+  | 'gpt-image-1.5'
+  | 'chatgpt-image-latest'
+  | 'dall-e-2'
+  | 'dall-e-3';
 
 /**
  * The response from the image generation endpoint.
@@ -553,10 +561,10 @@ export interface ImageEditParamsBase {
   /**
    * The image(s) to edit. Must be a supported image file or an array of images.
    *
-   * For the GPT image models (`gpt-image-1`, `gpt-image-1-mini`, and
-   * `gpt-image-1.5`), each image should be a `png`, `webp`, or `jpg` file less than
-   * 50MB. You can provide up to 16 images. `chatgpt-image-latest` follows the same
-   * input constraints as GPT image models.
+   * For the GPT image models (`gpt-image-1`, `gpt-image-1-mini`, `gpt-image-1.5`,
+   * `gpt-image-2`, `gpt-image-2-2026-04-21`, and `chatgpt-image-latest`), each image
+   * should be a `png`, `webp`, or `jpg` file less than 50MB. You can provide up to
+   * 16 images.
    *
    * For `dall-e-2`, you can only provide one image, and it should be a square `png`
    * file less than 4MB.
@@ -571,9 +579,14 @@ export interface ImageEditParamsBase {
 
   /**
    * Allows to set transparency for the background of the generated image(s). This
-   * parameter is only supported for the GPT image models. Must be one of
-   * `transparent`, `opaque` or `auto` (default value). When `auto` is used, the
-   * model will automatically determine the best background for the image.
+   * parameter is only supported for GPT image models that support transparent
+   * backgrounds. Must be one of `transparent`, `opaque`, or `auto` (default value).
+   * When `auto` is used, the model will automatically determine the best background
+   * for the image.
+   *
+   * `gpt-image-2` and `gpt-image-2-2026-04-21` do not support transparent
+   * backgrounds. Requests with `background` set to `transparent` will return an
+   * error for these models; use `opaque` or `auto` instead.
    *
    * If `transparent`, the output format needs to support transparency, so it should
    * be set to either `png` (default value) or `webp`.
@@ -597,7 +610,10 @@ export interface ImageEditParamsBase {
   mask?: Uploadable;
 
   /**
-   * The model to use for image generation. Defaults to `gpt-image-1.5`.
+   * The model to use for image generation. One of `dall-e-2` or a GPT image model
+   * (`gpt-image-1`, `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,
+   * `gpt-image-2-2026-04-21`, or `chatgpt-image-latest`). Defaults to
+   * `gpt-image-1.5`.
    */
   model?: (string & {}) | ImageModel | null;
 
@@ -645,11 +661,19 @@ export interface ImageEditParamsBase {
   response_format?: 'url' | 'b64_json' | null;
 
   /**
-   * The size of the generated images. Must be one of `1024x1024`, `1536x1024`
-   * (landscape), `1024x1536` (portrait), or `auto` (default value) for the GPT image
-   * models, and one of `256x256`, `512x512`, or `1024x1024` for `dall-e-2`.
+   * The size of the generated images. For `gpt-image-2` and
+   * `gpt-image-2-2026-04-21`, arbitrary resolutions are supported as `WIDTHxHEIGHT`
+   * strings, for example `1536x864`. Width and height must both be divisible by 16
+   * and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above
+   * `2560x1440` are experimental, and the maximum supported resolution is
+   * `3840x2160`. The requested size must also satisfy the model's current pixel and
+   * edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are
+   * supported by the GPT image models; `auto` is supported for models that allow
+   * automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or
+   * `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or
+   * `1024x1792`.
    */
-  size?: '256x256' | '512x512' | '1024x1024' | '1536x1024' | '1024x1536' | 'auto' | null;
+  size?: (string & {}) | '256x256' | '512x512' | '1024x1024' | '1536x1024' | '1024x1536' | 'auto' | null;
 
   /**
    * Edit the image in streaming mode. Defaults to `false`. See the
@@ -701,9 +725,14 @@ export interface ImageGenerateParamsBase {
 
   /**
    * Allows to set transparency for the background of the generated image(s). This
-   * parameter is only supported for the GPT image models. Must be one of
-   * `transparent`, `opaque` or `auto` (default value). When `auto` is used, the
-   * model will automatically determine the best background for the image.
+   * parameter is only supported for GPT image models that support transparent
+   * backgrounds. Must be one of `transparent`, `opaque`, or `auto` (default value).
+   * When `auto` is used, the model will automatically determine the best background
+   * for the image.
+   *
+   * `gpt-image-2` and `gpt-image-2-2026-04-21` do not support transparent
+   * backgrounds. Requests with `background` set to `transparent` will return an
+   * error for these models; use `opaque` or `auto` instead.
    *
    * If `transparent`, the output format needs to support transparency, so it should
    * be set to either `png` (default value) or `webp`.
@@ -712,8 +741,9 @@ export interface ImageGenerateParamsBase {
 
   /**
    * The model to use for image generation. One of `dall-e-2`, `dall-e-3`, or a GPT
-   * image model (`gpt-image-1`, `gpt-image-1-mini`, `gpt-image-1.5`). Defaults to
-   * `dall-e-2` unless a parameter specific to the GPT image models is used.
+   * image model (`gpt-image-1`, `gpt-image-1-mini`, `gpt-image-1.5`, `gpt-image-2`,
+   * or `gpt-image-2-2026-04-21`). Defaults to `dall-e-2` unless a parameter specific
+   * to the GPT image models is used.
    */
   model?: (string & {}) | ImageModel | null;
 
@@ -773,12 +803,20 @@ export interface ImageGenerateParamsBase {
   response_format?: 'url' | 'b64_json' | null;
 
   /**
-   * The size of the generated images. Must be one of `1024x1024`, `1536x1024`
-   * (landscape), `1024x1536` (portrait), or `auto` (default value) for the GPT image
-   * models, one of `256x256`, `512x512`, or `1024x1024` for `dall-e-2`, and one of
-   * `1024x1024`, `1792x1024`, or `1024x1792` for `dall-e-3`.
+   * The size of the generated images. For `gpt-image-2` and
+   * `gpt-image-2-2026-04-21`, arbitrary resolutions are supported as `WIDTHxHEIGHT`
+   * strings, for example `1536x864`. Width and height must both be divisible by 16
+   * and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above
+   * `2560x1440` are experimental, and the maximum supported resolution is
+   * `3840x2160`. The requested size must also satisfy the model's current pixel and
+   * edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are
+   * supported by the GPT image models; `auto` is supported for models that allow
+   * automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or
+   * `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or
+   * `1024x1792`.
    */
   size?:
+    | (string & {})
     | 'auto'
     | '1024x1024'
     | '1536x1024'

package/src/resources/realtime/api.md

@@ -56,6 +56,8 @@ Types:
 - <code><a href="./src/resources/realtime/realtime.ts">RealtimeMcpToolCall</a></code>
 - <code><a href="./src/resources/realtime/realtime.ts">RealtimeMcpToolExecutionError</a></code>
 - <code><a href="./src/resources/realtime/realtime.ts">RealtimeMcphttpError</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeReasoning</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeReasoningEffort</a></code>
 - <code><a href="./src/resources/realtime/realtime.ts">RealtimeResponse</a></code>
 - <code><a href="./src/resources/realtime/realtime.ts">RealtimeResponseCreateAudioOutput</a></code>
 - <code><a href="./src/resources/realtime/realtime.ts">RealtimeResponseCreateMcpTool</a></code>
@@ -75,6 +77,22 @@ Types:
 - <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranscriptionSessionAudioInput</a></code>
 - <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranscriptionSessionAudioInputTurnDetection</a></code>
 - <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranscriptionSessionCreateRequest</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationClientEvent</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationClientSecretCreateRequest</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationClientSecretCreateResponse</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationInputAudioBufferAppendEvent</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationInputTranscriptDeltaEvent</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationOutputAudioDeltaEvent</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationOutputTranscriptDeltaEvent</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationServerEvent</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationSession</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationSessionCloseEvent</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationSessionClosedEvent</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationSessionCreateRequest</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationSessionCreatedEvent</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationSessionUpdateEvent</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationSessionUpdateRequest</a></code>
+- <code><a href="./src/resources/realtime/realtime.ts">RealtimeTranslationSessionUpdatedEvent</a></code>
 - <code><a href="./src/resources/realtime/realtime.ts">RealtimeTruncation</a></code>
 - <code><a href="./src/resources/realtime/realtime.ts">RealtimeTruncationRetentionRatio</a></code>
 - <code><a href="./src/resources/realtime/realtime.ts">ResponseAudioDeltaEvent</a></code>
@@ -108,7 +126,6 @@ Types:
 
 Types:
 
-- <code><a href="./src/resources/realtime/client-secrets.ts">RealtimeSessionClientSecret</a></code>
 - <code><a href="./src/resources/realtime/client-secrets.ts">RealtimeSessionCreateResponse</a></code>
 - <code><a href="./src/resources/realtime/client-secrets.ts">RealtimeTranscriptionSessionCreateResponse</a></code>
 - <code><a href="./src/resources/realtime/client-secrets.ts">RealtimeTranscriptionSessionTurnDetection</a></code>

package/src/resources/realtime/calls.ts

@@ -134,6 +134,7 @@ export interface CallAcceptParams {
     | (string & {})
     | 'gpt-realtime'
     | 'gpt-realtime-1.5'
+    | 'gpt-realtime-2'
     | 'gpt-realtime-2025-08-28'
     | 'gpt-4o-realtime-preview'
     | 'gpt-4o-realtime-preview-2024-10-01'
@@ -157,12 +158,23 @@ export interface CallAcceptParams {
    */
   output_modalities?: Array<'text' | 'audio'>;
 
+  /**
+   * Whether the model may call multiple tools in parallel. Only supported by
+   * reasoning Realtime models such as `gpt-realtime-2`.
+   */
+  parallel_tool_calls?: boolean;
+
   /**
    * Reference to a prompt template and its variables.
    * [Learn more](https://platform.openai.com/docs/guides/text?api-mode=responses#reusable-prompts).
    */
   prompt?: ResponsesAPI.ResponsePrompt | null;
 
+  /**
+   * Configuration for reasoning-capable Realtime models such as `gpt-realtime-2`.
+   */
+  reasoning?: RealtimeAPI.RealtimeReasoning;
+
   /**
    * How the model chooses tools. Provide one of the string modes or force a specific
    * function/MCP tool.

package/src/resources/realtime/client-secrets.ts

@@ -41,32 +41,18 @@ export class ClientSecrets extends APIResource {
 }
 
 /**
- * Ephemeral key returned by the API.
+ * A Realtime session configuration object.
  */
-export interface RealtimeSessionClientSecret {
-  /**
-   * Timestamp for when the token expires. Currently, all tokens expire after one
-   * minute.
-   */
-  expires_at: number;
-
+export interface RealtimeSessionCreateResponse {
   /**
-   * Ephemeral key usable in client environments to authenticate connections to the
-   * Realtime API. Use this in client-side environments rather than a standard API
-   * token, which should only be used server-side.
+   * Unique identifier for the session that looks like `sess_1234567890abcdef`.
    */
-  value: string;
-}
+  id: string;
 
-/**
- * A new Realtime session configuration, with an ephemeral key. Default TTL for
- * keys is one minute.
- */
-export interface RealtimeSessionCreateResponse {
   /**
-   * Ephemeral key returned by the API.
+   * The object type. Always `realtime.session`.
    */
-  client_secret: RealtimeSessionClientSecret;
+  object: 'realtime.session';
 
   /**
    * The type of session to create. Always `realtime` for the Realtime API.
@@ -78,6 +64,11 @@ export interface RealtimeSessionCreateResponse {
    */
   audio?: RealtimeSessionCreateResponse.Audio;
 
+  /**
+   * Expiration timestamp for the session, in seconds since epoch.
+   */
+  expires_at?: number;
+
   /**
    * Additional fields to include in server outputs.
    *
@@ -115,6 +106,7 @@ export interface RealtimeSessionCreateResponse {
     | (string & {})
     | 'gpt-realtime'
     | 'gpt-realtime-1.5'
+    | 'gpt-realtime-2'
     | 'gpt-realtime-2025-08-28'
     | 'gpt-4o-realtime-preview'
     | 'gpt-4o-realtime-preview-2024-10-01'
@@ -144,6 +136,11 @@ export interface RealtimeSessionCreateResponse {
    */
   prompt?: ResponsesAPI.ResponsePrompt | null;
 
+  /**
+   * Configuration for reasoning-capable Realtime models such as `gpt-realtime-2`.
+   */
+  reasoning?: RealtimeAPI.RealtimeReasoning;
+
   /**
    * How the model chooses tools. Provide one of the string modes or force a specific
    * function/MCP tool.
@@ -215,16 +212,6 @@ export namespace RealtimeSessionCreateResponse {
        */
       noise_reduction?: Input.NoiseReduction;
 
-      /**
-       * Configuration for input audio transcription, defaults to off and can be set to
-       * `null` to turn off once on. Input audio transcription is not native to the
-       * model, since the model consumes audio directly. Transcription runs
-       * asynchronously through
-       * [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription)
-       * and should be treated as guidance of input audio content rather than precisely
-       * what the model heard. The client can optionally set the language and prompt for
-       * transcription, these offer additional guidance to the transcription service.
-       */
       transcription?: RealtimeAPI.AudioTranscription;
 
       /**
@@ -241,6 +228,9 @@ export namespace RealtimeSessionCreateResponse {
        * trails off with "uhhm", the model will score a low probability of turn end and
        * wait longer for the user to continue speaking. This can be useful for more
        * natural conversations, but may have a higher latency.
+       *
+       * For `gpt-realtime-whisper` transcription sessions, turn detection must be set to
+       * `null`; VAD is not supported.
        */
       turn_detection?: Input.ServerVad | Input.SemanticVad | null;
     }
@@ -640,17 +630,15 @@ export namespace RealtimeTranscriptionSessionCreateResponse {
        */
       noise_reduction?: Input.NoiseReduction;
 
-      /**
-       * Configuration of the transcription model.
-       */
       transcription?: RealtimeAPI.AudioTranscription;
 
       /**
        * Configuration for turn detection. Can be set to `null` to turn off. Server VAD
        * means that the model will detect the start and end of speech based on audio
-       * volume and respond at the end of user speech.
+       * volume and respond at the end of user speech. For `gpt-realtime-whisper`, this
+       * must be `null`; VAD is not supported.
        */
-      turn_detection?: ClientSecretsAPI.RealtimeTranscriptionSessionTurnDetection;
+      turn_detection?: ClientSecretsAPI.RealtimeTranscriptionSessionTurnDetection | null;
     }
 
     export namespace Input {
@@ -672,7 +660,8 @@ export namespace RealtimeTranscriptionSessionCreateResponse {
 /**
  * Configuration for turn detection. Can be set to `null` to turn off. Server VAD
  * means that the model will detect the start and end of speech based on audio
- * volume and respond at the end of user speech.
+ * volume and respond at the end of user speech. For `gpt-realtime-whisper`, this
+ * must be `null`; VAD is not supported.
  */
 export interface RealtimeTranscriptionSessionTurnDetection {
   /**
@@ -763,7 +752,6 @@ export namespace ClientSecretCreateParams {
 
 export declare namespace ClientSecrets {
   export {
-    type RealtimeSessionClientSecret as RealtimeSessionClientSecret,
     type RealtimeSessionCreateResponse as RealtimeSessionCreateResponse,
     type RealtimeTranscriptionSessionCreateResponse as RealtimeTranscriptionSessionCreateResponse,
     type RealtimeTranscriptionSessionTurnDetection as RealtimeTranscriptionSessionTurnDetection,

package/src/resources/realtime/index.ts

@@ -3,7 +3,6 @@
 export { Calls, type CallAcceptParams, type CallReferParams, type CallRejectParams } from './calls';
 export {
   ClientSecrets,
-  type RealtimeSessionClientSecret,
   type RealtimeSessionCreateResponse,
   type RealtimeTranscriptionSessionCreateResponse,
   type RealtimeTranscriptionSessionTurnDetection,