Wazza Engine LogoWazza Engine

Image Generation

Generate stunning images using state-of-the-art AI models from Black Forest Labs, Google, OpenAI, and Midjourney

Overview

Wazza Engine provides access to 67 image generation models across 5 providers:

Black Forest Labs (FLUX)

10 models including FLUX.1 schnell, dev, and specialized variants

Pricing: 1-3 credits per generation

Google Gemini (Imagen)

4 models: Imagen 4.0, Ultra, Fast, and Gemini 2.5 Flash

Pricing: 2-4 credits per generation

OpenAI

GPT Image 1 model

Pricing: 15 credits per generation

Midjourney

Midjourney v6 via ImagineAPI

Pricing: 4-8 credits per generation

Using Chat Orchestrator (Recommended)

The simplest way to generate images is using the Chat Orchestrator, powered by Claude 4.5 Sonnet:

import WazzaEngine from '@wazza/engine';

const wazza = new WazzaEngine({
  apiKey: process.env.WAZZA_API_KEY
});

// Natural language request - AI chooses the best model
const response = await wazza.chat({
  message: "Generate a photorealistic portrait of a woman with red hair",
  parseIntent: true
});

console.log('Job ID:', response.jobId);

// Wait for completion
const result = await wazza.waitForCompletion(response.jobId);
console.log('Image URL:', result.output.url);

Direct Model Selection

For more control, specify the exact model and provider:

FLUX Example

// Using FLUX.1 [dev] for high-quality generation
const response = await wazza.generate({
  provider: 'black-forest-labs',
  model: 'flux-dev',
  prompt: 'A serene mountain landscape at sunset, 8k quality',
  parameters: {
    width: 1024,
    height: 768,
    steps: 30,
    guidance: 7.5
  }
});

Imagen 4.0 Example

// Using Google Imagen 4.0 Ultra
const response = await wazza.generate({
  provider: 'google-gemini',
  model: 'imagen-4.0-ultra-generate-001',
  prompt: 'A futuristic cityscape with flying cars',
  parameters: {
    aspectRatio: '16:9',
    numberOfImages: 4
  }
});

Image-to-Image Generation

Transform existing images using FLUX or Topaz Labs models:

// Image-to-image with FLUX
const response = await wazza.generate({
  provider: 'black-forest-labs',
  model: 'flux-canny-dev',
  prompt: 'Transform into a watercolor painting',
  parameters: {
    image: 'https://example.com/input.jpg',
    strength: 0.75
  }
});

Inpainting

Edit specific areas of images using FLUX Fill:

// Inpainting with FLUX Fill
const response = await wazza.generate({
  provider: 'black-forest-labs',
  model: 'flux-fill-dev',
  prompt: 'Add a mountain range in the background',
  parameters: {
    image: 'https://example.com/original.jpg',
    mask: 'https://example.com/mask.png'
  }
});

Image Upscaling & Enhancement

Use Topaz Labs (42 models) for professional upscaling and enhancement:

// Upscale 4x with high quality
const response = await wazza.generate({
  provider: 'topaz-labs',
  model: 'ahq-12', // Apollo HQ 12
  parameters: {
    image: 'https://example.com/low-res.jpg',
    scale: 4
  }
});

Best Practices

1. Prompt Engineering

  • Be specific about style, lighting, and composition
  • Include quality keywords like "8k", "highly detailed", "professional"
  • Use negative prompts to avoid unwanted elements

2. Model Selection

  • FLUX schnell: Fast generation, lower quality
  • FLUX dev: High quality, slower generation
  • Imagen Ultra: Best for photorealism
  • Midjourney: Best for artistic styles

3. Cost Optimization

  • Use Chat Orchestrator to automatically select cost-effective models
  • Start with lower step counts for testing
  • Use FLUX schnell for rapid prototyping
Video Generation