Skip to main content
Version: 1.11.0-beta

generate

Initiates an image generation run or image edit using the ImagineoAI API. Supports multiple model types including Flux Dev, OpenAI, Google Imagen4, Flux Kontext Max, Flux Kontext Multi, WAN Image 2.1, WAN Image 2.2, Nanobanana, and Qwen Image Edit.

Signature:

client.images.generate(input: GenerateImageInput): Promise<GenerateImageOutput | null>
  • input: GenerateImageInput (see src/types.ts for full schema)
    • To generate a new image, provide a prompt and other parameters.
    • To edit an image, provide a reference_url (URL to the image to edit) along with your prompt and options.
    • For Google Imagen4, include aspect_ratio parameter (required).
    • For Flux Kontext Max, include aspect_ratio parameter (optional, defaults to "1:1").
    • For WAN Image 2.1, set model_type: "wan-image-2.1" and optionally adjust strength_model.
    • For WAN Image 2.2 (dual LoRA), set model_type: "wan-image-2.2" and provide both lora_low_name and lora_high_name.
    • For Qwen Image Edit, set model_type: "qwen-image-edit" with optional LoRA and strength parameters.
  • Returns: GenerateImageOutput or null if the API returns 204 No Content
  • Throws: On validation, network, or API error.

Examples​

Generate an image with Flux Dev (default)​

const client = new ImagineoAIClient(apiUrl, { apiKey: 'sk-...' });
const result = await client.images.generate({
  prompt: 'A cat in a spacesuit',
  model_type: 'flux-dev',  // Optional, this is the default
  width: 512,
  height: 512,
});
if (result) {
  console.log(result.image_url);
}

Generate an image with Google Imagen4​

const client = new ImagineoAIClient(apiUrl, { apiKey: 'sk-...' });
const result = await client.images.generate({
  prompt: 'A futuristic city at sunset',
  model_type: 'google-imagen4',
  aspect_ratio: '16:9',  // Required for Google Imagen4
  reference_image_url: 'https://example.com/style-reference.jpg' // Optional style reference
});
if (result) {
  console.log(result.image_url);
}

Generate an image with Flux Kontext Max​

const client = new ImagineoAIClient(apiUrl, { apiKey: 'sk-...' });
const result = await client.images.generate({
  prompt: 'A majestic dragon soaring through clouds at golden hour',
  model_type: 'flux-kontext-max',
  aspect_ratio: '16:9',  // Optional, defaults to "1:1"
  seed: 42,  // Optional, for reproducible results
});
if (result) {
  console.log(result.image_url);
}

Generate an image with Nanobanana​

const client = new ImagineoAIClient(apiUrl, { apiKey: 'sk-...' });
const result = await client.images.generate({
  prompt: 'A vibrant underwater scene with coral reefs',
  model_type: 'nanobanana',
  reference_image_url: 'https://example.com/style-guide.jpg'  // Optional reference for style
});
if (result) {
  console.log(result.image_url);
}

Edit an image using a reference URL​

const client = new ImagineoAIClient(apiUrl, { apiKey: 'sk-...' });
const result = await client.images.generate({
  prompt: 'Make the cat astronaut suit red',
  reference_image_url: 'https://example.com/mycat.png', // URL to the image to edit
  model_type: 'openai',  // or 'google-imagen4' with aspect_ratio
  width: 512,
  height: 512,
});
if (result) {
  console.log(result.image_url);
}

Generate with WAN Image 2.1​

const client = new ImagineoAIClient(apiUrl, { apiKey: 'sk-...' });
const result = await client.images.generate({
  prompt: 'A portrait in the style of the fine-tuned model',
  model_id: 'your-model-uuid',  // Required: Turso DB model ID
  model_type: 'wan-image-2.1',  // WAN Image 2.1 model
  width: 1920,
  height: 1088,
  strength_model: 1.5,  // Optional: Model strength (0-2, default: 1)
  model_weight: 0.7,  // Optional: Model weight (0-1)
  guidance: 7.5,  // Optional: Guidance scale
});
if (result) {
  console.log(result.image_url);
}

Generate with WAN Image 2.2 (Dual LoRA)​

const client = new ImagineoAIClient(apiUrl, { apiKey: 'sk-...' });
const result = await client.images.generate({
  prompt: 'A professional fashion photograph with natural lighting',
  model_id: 'your-model-uuid',  // Required: Turso DB model ID
  model_type: 'wan-image-2.2',  // WAN Image 2.2 model with dual LoRA
  width: 1200,
  height: 1500,
  lora_low_name: 'wan2.2/wan2.2-lora-instagirl-2.0/Instagirlv2.0_lownoise.safetensors',  // Required
  lora_low_strength: 0.5,  // Optional: Low LoRA strength (0-2, default: 0.5)
  lora_high_name: 'wan2.2/wan2.2-lora-instagirl-2.0/Instagirlv2.0_hinoise.safetensors',  // Required
  lora_high_strength: 1.0,  // Optional: High LoRA strength (0-2, default: 1.0)
  batch_size: 1,  // Optional: Number of images to generate (1-4, default: 1)
  model_weight: 0.7,  // Optional: Model weight (0-1)
  guidance: 7.5,  // Optional: Guidance scale
});
if (result) {
  console.log(result.image_url);
}

Generate with Qwen Image Edit​

const client = new ImagineoAIClient(apiUrl, { apiKey: 'sk-...' });
const result = await client.images.generate({
  prompt: 'Transform the scene into a fantasy landscape with magical elements',
  model_type: 'qwen-image-edit',
  lora_1: 'fantasy-style-v1',  // Optional: LoRA model for style
  strength_model: 1.5,  // Optional: LoRA strength (0-2, default: 1)
  width: 1024,
  height: 1024,
});
if (result) {
  console.log(result.image_url);
}

Model-Specific Notes​

Flux Dev Model​

  • Type: flux-dev
  • Default model if model_type is not specified
  • High-quality image generation with Flux architecture
  • Uses width and height parameters for dimensions

OpenAI Model​

  • Type: openai
  • OpenAI's DALL-E image generation model
  • Supports both image generation and editing
  • Optional reference_image_url for style reference

Google Imagen4​

  • Requires aspect_ratio parameter (e.g., '16:9', '1:1', '9:16')
  • Supports style transfer via reference_image_url
  • Generates high-quality images with advanced capabilities
  • Optimized for specific aspect ratios

Flux Kontext Max​

  • Type: flux-kontext-max
  • Description: Advanced image generation model with superior quality and detail
  • Required Parameters:
    • prompt: string - The text prompt for image generation
  • Optional Parameters:
    • aspect_ratio?: string - Aspect ratio of the generated image (defaults to "1:1")
      • Supported ratios: "1:1", "16:9", "9:16", "4:3", "3:4", "3:2", "2:3", "4:5", "5:4", "21:9", "9:21", "2:1", "1:2"
    • seed?: number - Random seed for reproducible generation (integer ≥ 0)

Nanobanana​

  • Type: nanobanana
  • Description: Replicate-based model for advanced image generation, combination, and editing
  • Required Parameters:
    • prompt: string - The text prompt for image generation
  • Optional Parameters:
    • reference_image_url?: string - URL of reference image for style guidance
  • Features:
    • Supports style transfer through reference images
    • Can combine multiple images (see combine method)
    • Supports image editing (see edit.nanobanana methods)
    • Handles both JSON and FormData input formats
    • Optimized for both artistic and photorealistic outputs

WAN Image 2.1​

  • Type: wan-image-2.1
  • Description: WAN Image generation model version 2.1
  • Required Parameters:
    • prompt: string - The text prompt for image generation
    • model_id: string - UUID of the model in Turso database
    • model_type: "wan-image-2.1" - Selects the WAN Image 2.1 model
    • width: number - Image width (64-2048)
    • height: number - Image height (64-2048)
  • Optional Parameters:
    • strength_model?: number - Model strength (0-2, default: 1)
    • model_weight?: number - Model weight (0-1, default: 0.7)
    • guidance?: number - Guidance scale (0-100, default: 7.5)
  • Features:
    • Optimized for fine-tuned models and LoRAs
    • Batch size of 2 for better quality
    • Uses specialized ComfyDeploy deployment
    • Maintains compatibility with existing model permissions

WAN Image 2.2 (Dual LoRA)​

  • Type: wan-image-2.2
  • Description: Latest WAN Image model with dual LoRA support
  • Required Parameters:
    • prompt: string - The text prompt
    • model_id: string - UUID of the model
    • lora_low_name: string - Low LoRA model name
    • lora_high_name: string - High LoRA model name
  • Optional Parameters:
    • lora_low_strength?: number - Low LoRA strength (0-2, default: 0.5)
    • lora_high_strength?: number - High LoRA strength (0-2, default: 1.0)

Flux Kontext Multi​

  • Type: flux-kontext-multi
  • Description: Multi-image combination and generation model
  • Combines multiple input images with advanced style transfer

Qwen Image Edit​

  • Type: qwen-image-edit
  • Description: Advanced text-based image editing model powered by ComfyDeploy
  • Required Parameters:
    • prompt: string - The text prompt describing the desired edit
    • width: number - Image width (64-2048)
    • height: number - Image height (64-2048)
  • Optional Parameters:
    • lora_1?: string - Optional LoRA model for style enhancement
    • strength_model?: number - LoRA strength (0-2, default: 1)
    • negative_prompt?: string - Text describing what to avoid in the generation
  • Features:
    • AI-powered image editing with text prompts
    • Support for negative prompts for precise control
    • LoRA model integration for style customization
    • Seamless integration with existing generate workflow
    • No breaking changes to existing API

Available Model Types​

The SDK now uses a unified set of model types:

  • flux-dev - Default Flux Dev model
  • wan-image-2.1 - WAN Image version 2.1
  • wan-image-2.2 - WAN Image version 2.2 with dual LoRA
  • flux-kontext-multi - Multi-image combination
  • flux-kontext-max - Advanced high-quality generation
  • google-imagen4 - Google's Imagen 4 model
  • openai - OpenAI's DALL-E model
  • nanobanana - Replicate-based model for generation and editing
  • qwen-image-edit - Advanced text-based image editing

General Notes​

  • The workflow_type parameter is deprecated. Use model_type instead.
  • The method branches based on the model_type parameter to use the appropriate generation backend.
  • SDK always sends data as JSON—all files/images must be provided as URLs, not file uploads.
  • Input is validated with Zod (GenerateImageInputSchema).
  • Makes a POST request to /api/v1/runs/generate.
  • See the API Reference for details on input/output types.