Image Model API Reference

public class VertexAiImageModel implements ImageModel {
    public Response<Image> generate(String prompt);
    public Response<List<Image>> generate(String prompt, int n);
    public Response<Image> edit(Image image, String prompt);
    public Response<Image> edit(Image image, Image mask, String prompt);
    public static Builder builder();
}

/**
 * Generate a single image from a text prompt.
 *
 * @param prompt The text description of the desired image
 * @return Response containing the generated image
 */
public Response<Image> generate(String prompt);

/**
 * Generate multiple image variations from a text prompt.
 *
 * @param prompt The text description of the desired images
 * @param n The number of image variations to generate
 * @return Response containing list of generated images
 */
public Response<List<Image>> generate(String prompt, int n);

/**
 * Edit an image using a text prompt.
 *
 * @param image The source image to edit
 * @param prompt The text description of desired changes
 * @return Response containing the edited image
 */
public Response<Image> edit(Image image, String prompt);

/**
 * Edit an image using a mask and text prompt.
 *
 * @param image The source image to edit
 * @param mask The mask image indicating areas to modify (white = edit, black = preserve)
 * @param prompt The text description of desired changes for masked areas
 * @return Response containing the edited image
 */
public Response<Image> edit(Image image, Image mask, String prompt);

/**
 * Create a new builder for configuring a VertexAiImageModel.
 *
 * @return A new Builder instance
 */
public static Builder builder();

public Builder endpoint(String endpoint);
public Builder project(String project);
public Builder location(String location);
public Builder publisher(String publisher);
public Builder modelName(String modelName);

public Builder seed(Long seed);                              // Random seed (0-4,294,967,295)
public Builder language(String language);                    // Language code
public Builder guidanceScale(Integer guidanceScale);         // Edit strength (0-21+)
public Builder negativePrompt(String negativePrompt);        // What to avoid
public Builder sampleImageStyle(ImageStyle sampleImageStyle); // Style (Imagen v1 only)
public Builder sampleImageSize(Integer sampleImageSize);     // Output size in pixels
public Builder aspectRatio(AspectRatio aspectRatio);         // Image aspect ratio
public Builder mimeType(MimeType mimeType);                  // PNG or JPEG
public Builder compressionQuality(Integer compressionQuality); // JPEG quality (1-100)
public Builder personGeneration(PersonGeneration personGeneration); // Person policy
public Builder watermark(Boolean watermark);                 // Add watermark
public Builder maxRetries(Integer maxRetries);               // Retry attempts (default: 3)
public Builder persistToCloudStorage(String gcsUri);         // Save to GCS
public Builder withPersisting();                             // Enable local persistence
public Builder persistTo(Path persistTo);                    // Local save path
public Builder logRequests(Boolean logRequests);             // Log requests
public Builder logResponses(Boolean logResponses);           // Log responses
public VertexAiImageModel build();

/**
 * Create a VertexAiImageModel with explicit parameters.
 */
public VertexAiImageModel(
    String endpoint,
    String project,
    String location,
    String publisher,
    String modelName,
    Long seed,
    String language,
    Integer guidanceScale,
    String negativePrompt,
    ImageStyle sampleImageStyle,
    Integer sampleImageSize,
    AspectRatio aspectRatio,
    PersonGeneration personGeneration,
    Integer maxRetries,
    MimeType mimeType,
    Integer compressionQuality,
    Boolean addWatermark,
    String cloudStorageBucket,
    Boolean withPersisting,
    Path persistTo,
    Boolean logRequests,
    Boolean logResponses
);

public enum AspectRatio {
    SQUARE,         // 1:1 - toString() returns "1:1"
    PORTRAIT,       // 9:16 - toString() returns "9:16"
    LANDSCAPE,      // 16:9 - toString() returns "16:9"
    THREE_FOURTHS,  // 3:4 - toString() returns "3:4"
    FOUR_THIRDS;    // 4:3 - toString() returns "4:3"

    public String toString();
}

public enum ImageStyle {
    PHOTOGRAPH,     // Photographic - toString() returns "photograph"
    DIGITAL_ART,    // Digital art - toString() returns "digital_art"
    LANDSCAPE,      // Landscape - toString() returns "landscape"
    SKETCH,         // Sketch - toString() returns "sketch"
    WATERCOLOR,     // Watercolor - toString() returns "watercolor"
    CYBERPUNK,      // Cyberpunk - toString() returns "cyberpunk"
    POP_ART;        // Pop art - toString() returns "pop_art"

    public String toString();
}

public enum MimeType {
    PNG,   // PNG format - toString() returns "image/png"
    JPEG;  // JPEG format - toString() returns "image/jpeg"

    public String toString();
}

public enum PersonGeneration {
    DONT_ALLOW,    // Don't generate people - toString() returns "dont_allow"
    ALLOW_ADULT,   // Allow adult only - toString() returns "allow_adult"
    ALLOW_ALL;     // Allow all - toString() returns "allow_all"

    public String toString();
}

public class Response<T> {
    public T content();
    public TokenUsage tokenUsage();
    public FinishReason finishReason();
}

public class Image {
    public byte[] bytes();
    public String base64Data();
    public URI uri();
}

// From bytes
public static Image fromBytes(byte[] bytes);

// From file
Image image = Image.fromBytes(Files.readAllBytes(Path.of("image.png")));

// From URL
Image image = Image.fromUrl(new URL("https://..."));