Image Generation

Generate AI images using Google's Imagen models through the client.images API.

Image generation requires reCAPTCHA configuration. See reCAPTCHA Configuration for setup instructions.

Upload Image

Upload an image for use as a reference in generation:

const result = await client.images.upload({
  imageBase64: 'base64-encoded-image-data',
  sessionId: GLabsClient.generateSessionId(),
  aspectRatio: '16:9', // Optional: '16:9' | '9:16' | '1:1'
});
 
console.log(result.mediaId);           // Use this ID as a reference
console.log(result.mediaGenerationId); // Same as mediaId
console.log(result.width);             // Image width
console.log(result.height);            // Image height

Upload Options

Option	Type	Required	Description
`imageBase64`	`string`	Yes	Base64-encoded image (with or without data URL prefix)
`sessionId`	`string`	Yes	Unique session identifier
`aspectRatio`	`AspectRatio`	No	Target aspect ratio

Generate Images

Generate images from a text prompt:

const result = await client.images.generate({
  prompt: 'A majestic mountain landscape at sunset',
  sessionId: GLabsClient.generateSessionId(),
  aspectRatio: '16:9',
  count: 4,         // Generate 4 images
  model: 'nanobanana', // or 'nanobananapro'
});
 
for (const image of result.images) {
  console.log(image.encodedImage); // Base64 image data
  console.log(image.mediaId);      // Media ID for video generation
  console.log(image.fifeUrl);      // CDN URL if available
}

Generate Options

Option	Type	Required	Description
`prompt`	`string`	Yes	Description of the image to generate
`sessionId`	`string`	Yes	Unique session identifier
`aspectRatio`	`AspectRatio`	Yes	Image aspect ratio
`projectId`	`string`	No*	Project ID (auto-resolves if not provided)
`references`	`ImageReference[]`	No	Reference images for style
`seed`	`number`	No	Random seed for reproducibility
`count`	`number`	No	Number of images (1-4, default: 1)
`model`	`ImageModel`	No	Model to use
`prompts`	`string[]`	No	Different prompts for each image

* Auto-resolves to first available project if not set in client config

Using Reference Images

Generate images inspired by uploaded references:

// First, upload reference images
const ref1 = await client.images.upload({
  imageBase64: styleImage,
  sessionId: GLabsClient.generateSessionId(),
});
 
// Generate using reference
const result = await client.images.generate({
  prompt: 'A similar landscape in winter',
  sessionId: GLabsClient.generateSessionId(),
  aspectRatio: '16:9',
  references: [
    { mediaId: ref1.mediaId },
  ],
});

Multiple Different Prompts

Generate images with different prompts in a single request:

const result = await client.images.generate({
  prompt: 'fallback prompt', // Used if prompts array is short
  sessionId: GLabsClient.generateSessionId(),
  aspectRatio: '16:9',
  prompts: [
    'A cat sitting on a windowsill',
    'A dog playing in the park',
    'A bird flying over the ocean',
    'A fish swimming in a coral reef',
  ],
});
 
// result.images[0] -> cat
// result.images[1] -> dog
// result.images[2] -> bird
// result.images[3] -> fish

Check Credit Status

Check your account's credit balance:

const credits = await client.images.getCreditStatus();
 
console.log(credits.credits);                    // Remaining credits
console.log(credits.userPaygateTier);            // Payment tier
console.log(credits.g1MembershipState);          // Membership state
console.log(credits.isUserAnimateCountryEnabled); // Animate feature access

Image Models

Model	Description
`nanobanana`	Standard Imagen model (GEM_PIX)
`nanobananapro`	Enhanced Imagen model (GEM_PIX_2)

Aspect Ratios

Ratio	Description
`16:9`	Landscape (default)
`9:16`	Portrait
`1:1`	Square

Error Handling

import { GLabsError } from '@getvrex/glabs-sdk';
 
try {
  const result = await client.images.generate({
    prompt: 'test',
    sessionId: GLabsClient.generateSessionId(),
    aspectRatio: '16:9',
  });
} catch (error) {
  if (error instanceof GLabsError) {
    switch (error.code) {
      case 'INVALID_ARGUMENT':
        console.error('Content violates policy');
        break;
      case 'HIGH_TRAFFIC':
        console.error('Service busy, try again later');
        break;
      case 'RECAPTCHA_REQUIRED':
        console.error('Configure reCAPTCHA provider');
        break;
      case 'RECAPTCHA_EVAL_FAILED':
        console.error('reCAPTCHA evaluation failed after retries');
        break;
      default:
        console.error('Error:', error.message);
    }
  }
}

Image Generation

On this page