@google/genai 1.14.0 → 1.15.0

package/dist/web/web.d.ts

@@ -1885,6 +1885,14 @@ export declare interface EnterpriseWebSearch {
     excludeDomains?: string[];
 }
 
+/** An entity representing the segmented area. */
+export declare interface EntityLabel {
+    /** The label of the segmented entity. */
+    label?: string;
+    /** The confidence score of the detected label. */
+    score?: number;
+}
+
 /** The environment being operated. */
 export declare enum Environment {
     /**
@@ -2702,6 +2710,14 @@ export declare interface GeneratedImage {
     enhancedPrompt?: string;
 }
 
+/** A generated image mask. */
+export declare interface GeneratedImageMask {
+    /** The generated image mask. */
+    mask?: Image_2;
+    /** The detected entities on the segmented area. */
+    labels?: EntityLabel[];
+}
+
 /** A generated video. */
 export declare interface GeneratedVideo {
     /** The output video */
@@ -2840,6 +2856,12 @@ export declare interface GenerateVideosConfig {
     generateAudio?: boolean;
     /** Image to use as the last frame of generated videos. Only supported for image to video use cases. */
     lastFrame?: Image_2;
+    /** The images to use as the references to generate the videos.
+     If this field is provided, the text prompt field must also be provided.
+     The image, video, or last_frame field are not supported. Each image must
+     be associated with a type. Veo 2 supports up to 3 asset images *or* 1
+     style image. */
+    referenceImages?: VideoGenerationReferenceImage[];
     /** Compression quality of the generated videos. */
     compressionQuality?: VideoCompressionQuality;
 }
@@ -5132,6 +5154,27 @@ export declare class Models extends BaseModule {
      * ```
      */
     recontextImage(params: types.RecontextImageParameters): Promise<types.RecontextImageResponse>;
+    /**
+     * Segments an image, creating a mask of a specified area.
+     *
+     * @param params - The parameters for segmenting an image.
+     * @return The response from the API.
+     *
+     * @example
+     * ```ts
+     * const response = await ai.models.segmentImage({
+     *  model: 'image-segmentation-001',
+     *  source: {
+     *    image: image,
+     *  },
+     *  config: {
+     *    mode: 'foreground',
+     *  },
+     * });
+     * console.log(response?.generatedMasks?.[0]?.mask?.imageBytes);
+     * ```
+     */
+    segmentImage(params: types.SegmentImageParameters): Promise<types.SegmentImageResponse>;
     /**
      * Fetches information about a model by name.
      *
@@ -5974,6 +6017,12 @@ export declare interface Schema {
 
 export declare type SchemaUnion = Schema | unknown;
 
+/** An image mask representing a brush scribble. */
+export declare interface ScribbleImage {
+    /** The brush scribble to guide segmentation. Valid for the interactive mode. */
+    image?: Image_2;
+}
+
 /** Google search entry point. */
 export declare interface SearchEntryPoint {
     /** Optional. Web content snippet that can be embedded in a web page or an app webview. */
@@ -5995,6 +6044,75 @@ export declare interface Segment {
     text?: string;
 }
 
+/** Configuration for segmenting an image. */
+export declare interface SegmentImageConfig {
+    /** Used to override HTTP request options. */
+    httpOptions?: HttpOptions;
+    /** Abort signal which can be used to cancel the request.
+
+     NOTE: AbortSignal is a client-only operation. Using it to cancel an
+     operation will not cancel the request in the service. You will still
+     be charged usage for any applicable operations.
+     */
+    abortSignal?: AbortSignal;
+    /** The segmentation mode to use. */
+    mode?: SegmentMode;
+    /** The maximum number of predictions to return up to, by top
+     confidence score. */
+    maxPredictions?: number;
+    /** The confidence score threshold for the detections as a decimal
+     value. Only predictions with a confidence score higher than this
+     threshold will be returned. */
+    confidenceThreshold?: number;
+    /** A decimal value representing how much dilation to apply to the
+     masks. 0 for no dilation. 1.0 means the masked area covers the whole
+     image. */
+    maskDilation?: number;
+    /** The binary color threshold to apply to the masks. The threshold
+     can be set to a decimal value between 0 and 255 non-inclusive.
+     Set to -1 for no binary color thresholding. */
+    binaryColorThreshold?: number;
+}
+
+/** The parameters for segmenting an image. */
+export declare interface SegmentImageParameters {
+    /** ID of the model to use. For a list of models, see `Google models
+     <https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models>`_. */
+    model: string;
+    /** A set of source input(s) for image segmentation. */
+    source: SegmentImageSource;
+    /** Configuration for image segmentation. */
+    config?: SegmentImageConfig;
+}
+
+/** The output images response. */
+export declare class SegmentImageResponse {
+    /** List of generated image masks.
+     */
+    generatedMasks?: GeneratedImageMask[];
+}
+
+/** A set of source input(s) for image segmentation. */
+export declare interface SegmentImageSource {
+    /** A text prompt for guiding the model during image segmentation.
+     Required for prompt mode and semantic mode, disallowed for other modes. */
+    prompt?: string;
+    /** The image to be segmented. */
+    image?: Image_2;
+    /** The brush scribble to guide segmentation.
+     Required for the interactive mode, disallowed for other modes. */
+    scribbleImage?: ScribbleImage;
+}
+
+/** Enum that represents the segmentation mode. */
+export declare enum SegmentMode {
+    FOREGROUND = "FOREGROUND",
+    BACKGROUND = "BACKGROUND",
+    PROMPT = "PROMPT",
+    SEMANTIC = "SEMANTIC",
+    INTERACTIVE = "INTERACTIVE"
+}
+
 /** Parameters for sending a message within a chat session.
 
  These parameters are used with the `chat.sendMessage()` method.
@@ -6882,6 +7000,7 @@ declare namespace types {
         ControlReferenceType,
         SubjectReferenceType,
         EditMode,
+        SegmentMode,
         VideoCompressionQuality,
         FileState,
         FileSource,
@@ -7012,6 +7131,13 @@ declare namespace types {
         RecontextImageConfig,
         RecontextImageParameters,
         RecontextImageResponse,
+        ScribbleImage,
+        SegmentImageSource,
+        SegmentImageConfig,
+        SegmentImageParameters,
+        EntityLabel,
+        GeneratedImageMask,
+        SegmentImageResponse,
         GetModelConfig,
         GetModelParameters,
         Endpoint,
@@ -7036,6 +7162,7 @@ declare namespace types {
         TokensInfo,
         ComputeTokensResponse,
         Video,
+        VideoGenerationReferenceImage,
         GenerateVideosConfig,
         GenerateVideosParameters,
         GeneratedVideo,
@@ -7500,6 +7627,17 @@ export declare enum VideoCompressionQuality {
     LOSSLESS = "LOSSLESS"
 }
 
+/** A reference image for video generation. */
+export declare interface VideoGenerationReferenceImage {
+    /** The reference image.
+     */
+    image?: Image_2;
+    /** The type of the reference image, which defines how the reference
+     image will be used to generate the video. Supported values are 'asset'
+     or 'style'. */
+    referenceType?: string;
+}
+
 /** Describes how the video in the Part should be used by the model. */
 export declare interface VideoMetadata {
     /** The frame rate of the video sent to the model. If not specified, the

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@google/genai",
-  "version": "1.14.0",
+  "version": "1.15.0",
   "description": "",
   "type": "module",
   "main": "dist/node/index.mjs",