Lumina-Image-2.0 is a 2 billion parameter flow-based diffusion transforer which features improved performance in image quality, typography, complex prompt understanding, and resource-efficiency.

fal-ai/lumina-image/v2

Lumina-Image-2.0 is a 2 billion parameter flow-based diffusion transforer which features improved performance in image quality, typography, complex prompt understanding, and resource-efficiency.

Inference

Commercial use

Streaming

Schema

LLMs

Playground API

SandboxOpen in Sandbox

Input

Prompt*

Additional Settings

Customize your input with more control.

Result

Idle

What would you like to do next?

fal-ai/lumina-image/v2

Schema

LLMs

Input

Prompt*

Additional Settings

Customize your input with more control.

Result

Idle

What would you like to do next?

Input

Prompt*

A vibrant and artistic digital composition featuring colorful splashes of paint in the background, creating an energetic and dynamic effect. The text 'Lumina on Fal' is elegantly integrated into the scene, standing out with a modern, bold, and slightly futuristic font. The colors are bright and varied, including neon blues, purples, pinks, and oranges, blending seamlessly in a fluid, abstract style. The text appears slightly illuminated, complementing the vivid splashes around it.

Additional Settings

Customize your input with more control.

Your request will cost $0.075 per megapixel.

Streaming

or try similar models free in Sandbox

Result

Idle

What would you like to do next?

Download

{
  "images": [
    {
      "height": 768,
      "width": 1024,
      "content_type": "image/jpeg",
      "url": "https://v3.fal.media/files/rabbit/pBwaEZysJhnstKWEHGpLc.png"
    }
  ],
  "prompt": ""
}

Your request will cost $0.075 per megapixel.

Readme

Lumina Image 2.0 - Advanced Flow-Based Text-to-Image Generation

Transform your creative vision into stunning images with Lumina Image 2.0, a powerful 2 billion parameter flow-based diffusion transformer built for developers who need reliable, high-quality results at scale.

Overview

Lumina Image 2.0 is a state-of-the-art text-to-image generation model that features improved performance in image quality, typography, complex prompt understanding, and resource-efficiency. Whether you're building a creative tool, content generation platform, or enhancing your application's visual features, Lumina Image provides the foundation you need with its unified framework approach.

Key Features

Transform natural language descriptions into detailed, high-quality images with advanced controls:

2 billion parameter flow-based diffusion transformer architecture
Enhanced typography and text rendering capabilities
Commercial usage rights included (Apache 2.0 license)
Support for resolutions up to 1024x1024
RESTful API with comprehensive SDKs
Streaming support for real-time generation feedback

Getting Started

Getting up and running with Lumina Image takes just a few minutes. Here's how to begin:

Install the SDK for your preferred language:

For JavaScript/TypeScript:

bash
npm install --save @fal-ai/client

For Python:

bash
pip install fal-client

Configure your authentication:

javascript
import { fal } from "@fal-ai/client";

fal.config({
  credentials: "YOUR_FAL_KEY_HERE"
});

Generate your first image:

javascript
const result = await fal.subscribe("fal-ai/lumina-image/v2", {
  input: {
    prompt: "A serene mountain landscape at sunset with snow-capped peaks"
  }
});

console.log(result.images[0].url);

Integration Examples

Here's a practical example of integrating Lumina Image into a web application:

javascript
import { fal } from "@fal-ai/client";

async function generateImage(prompt) {
  try {
    const result = await fal.subscribe("fal-ai/lumina-image/v2", {
      input: {
        prompt,
        height: 1024,
        width: 1024,
        guidance_scale: 4.0,
        num_inference_steps: 50
      }
    });
    return result.images[0];
  } catch (error) {
    console.error("Image generation failed:", error);
    throw error;
  }
}

API Parameters

prompt (required): Text description of the image to generate
height: Output image height (default: 1024)
width: Output image width (default: 1024)
guidance_scale: Strength of prompt adherence (default: 4.0)
num_inference_steps: Number of denoising steps (default: 50)
cfg_trunc_ratio: CFG truncation ratio (default: 0.25)
cfg_normalization: Enable CFG normalization (default: true)
generator: Random seed for reproducible results

Best Practices

Maximize the quality of your generated images by following these guidelines:

Write clear, detailed prompts that specify both content and style
Include artistic references when seeking specific visual outcomes
Use the Gemma-2-2B text encoder's capabilities for complex prompt understanding
Implement proper error handling and retry logic
Consider using streaming for better user experience

Advanced Usage

For more complex use cases, Lumina Image supports advanced parameters:

javascript
const advancedGeneration = await fal.subscribe("fal-ai/lumina-image/v2", {
  input: {
    prompt: "Professional portrait photography, golden hour lighting",
    guidance_scale: 4.0,
    num_inference_steps: 50,
    cfg_trunc_ratio: 0.25,
    cfg_normalization: true,
    generator: 42 // For reproducible results
  }
});

Technical Architecture

Core Components:

Model: 2B parameter Flow-based Large Diffusion Transformer (Flag-DiT)
Text Encoder: Gemma-2-2B for enhanced prompt understanding
VAE: FLUX-VAE-16CH variational autoencoder
Framework: Unified architecture treating text and image tokens jointly

Pricing

Cost: $0.075 per megapixel
Transparent usage-based pricing
No minimum commitments

Queue Management

For asynchronous processing:

javascript
// Submit request
const { request_id } = await fal.queue.submit("fal-ai/lumina-image/v2", {
  input: {
    prompt: "Your detailed image description"
  }
});

// Check status
const status = await fal.queue.status("fal-ai/lumina-image/v2", {
  requestId: request_id
});

// Get result
const result = await fal.queue.result("fal-ai/lumina-image/v2", {
  requestId: request_id
});

Error Handling

Implement robust error handling to ensure a smooth user experience:

javascript
try {
  const result = await generateImage(prompt);
  displayImage(result);
} catch (error) {
  if (error.status === 429) {
    // Handle rate limiting
    await handleRateLimit();
  } else {
    showErrorMessage("Failed to generate image. Please try again.");
  }
}

Technical Support

Our documentation is continuously updated with new examples and best practices. For additional support:

Visit our comprehensive API documentation
Join our developer community
Contact our technical support team
Monitor system status at status.fal.ai

Getting Started Today

Sign up for a fal.ai account
Generate your API key
Install the SDK
Make your first API call

Start building with Lumina Image 2.0 today and bring your creative vision to life through the power of advanced AI-generated imagery.

Readme

Lumina Image 2.0 - Advanced Flow-Based Text-to-Image Generation

Overview

Key Features

Transform natural language descriptions into detailed, high-quality images with advanced controls:

2 billion parameter flow-based diffusion transformer architecture
Enhanced typography and text rendering capabilities
Commercial usage rights included (Apache 2.0 license)
Support for resolutions up to 1024x1024
RESTful API with comprehensive SDKs
Streaming support for real-time generation feedback

Getting Started

Getting up and running with Lumina Image takes just a few minutes. Here's how to begin:

Install the SDK for your preferred language:

For JavaScript/TypeScript:

bash
npm install --save @fal-ai/client

For Python:

bash
pip install fal-client

Configure your authentication:

javascript
import { fal } from "@fal-ai/client";

fal.config({
  credentials: "YOUR_FAL_KEY_HERE"
});

Generate your first image:

javascript
const result = await fal.subscribe("fal-ai/lumina-image/v2", {
  input: {
    prompt: "A serene mountain landscape at sunset with snow-capped peaks"
  }
});

console.log(result.images[0].url);

Integration Examples

Here's a practical example of integrating Lumina Image into a web application:

javascript
import { fal } from "@fal-ai/client";

async function generateImage(prompt) {
  try {
    const result = await fal.subscribe("fal-ai/lumina-image/v2", {
      input: {
        prompt,
        height: 1024,
        width: 1024,
        guidance_scale: 4.0,
        num_inference_steps: 50
      }
    });
    return result.images[0];
  } catch (error) {
    console.error("Image generation failed:", error);
    throw error;
  }
}

API Parameters

prompt (required): Text description of the image to generate
height: Output image height (default: 1024)
width: Output image width (default: 1024)
guidance_scale: Strength of prompt adherence (default: 4.0)
num_inference_steps: Number of denoising steps (default: 50)
cfg_trunc_ratio: CFG truncation ratio (default: 0.25)
cfg_normalization: Enable CFG normalization (default: true)
generator: Random seed for reproducible results

Best Practices

Maximize the quality of your generated images by following these guidelines:

Write clear, detailed prompts that specify both content and style
Include artistic references when seeking specific visual outcomes
Use the Gemma-2-2B text encoder's capabilities for complex prompt understanding
Implement proper error handling and retry logic
Consider using streaming for better user experience

Advanced Usage

For more complex use cases, Lumina Image supports advanced parameters:

javascript
const advancedGeneration = await fal.subscribe("fal-ai/lumina-image/v2", {
  input: {
    prompt: "Professional portrait photography, golden hour lighting",
    guidance_scale: 4.0,
    num_inference_steps: 50,
    cfg_trunc_ratio: 0.25,
    cfg_normalization: true,
    generator: 42 // For reproducible results
  }
});

Technical Architecture

Core Components:

Model: 2B parameter Flow-based Large Diffusion Transformer (Flag-DiT)
Text Encoder: Gemma-2-2B for enhanced prompt understanding
VAE: FLUX-VAE-16CH variational autoencoder
Framework: Unified architecture treating text and image tokens jointly

Pricing

Cost: $0.075 per megapixel
Transparent usage-based pricing
No minimum commitments

Queue Management

For asynchronous processing:

javascript
// Submit request
const { request_id } = await fal.queue.submit("fal-ai/lumina-image/v2", {
  input: {
    prompt: "Your detailed image description"
  }
});

// Check status
const status = await fal.queue.status("fal-ai/lumina-image/v2", {
  requestId: request_id
});

// Get result
const result = await fal.queue.result("fal-ai/lumina-image/v2", {
  requestId: request_id
});

Error Handling

Implement robust error handling to ensure a smooth user experience:

javascript
try {
  const result = await generateImage(prompt);
  displayImage(result);
} catch (error) {
  if (error.status === 429) {
    // Handle rate limiting
    await handleRateLimit();
  } else {
    showErrorMessage("Failed to generate image. Please try again.");
  }
}

Technical Support

Our documentation is continuously updated with new examples and best practices. For additional support:

Visit our comprehensive API documentation
Join our developer community
Contact our technical support team
Monitor system status at status.fal.ai

Getting Started Today

Sign up for a fal.ai account
Generate your API key
Install the SDK
Make your first API call

Start building with Lumina Image 2.0 today and bring your creative vision to life through the power of advanced AI-generated imagery.