> ## Documentation Index
> Fetch the complete documentation index at: https://docs.remade.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Image Generation

> Learn how to Generate stunning images using Remade Canvas's powerful AI models and advanced features. 

## Generating Text to Image

To generate an image from a text prompt:

1. **Access the generation toolbar** - Click on the generation icon (stars icon) in the left toolbar to make the generation toolbar appear
2. **Write your prompt** - Enter a descriptive text prompt in the input field
3. **Enhance your prompt (optional)** - Click the "Enhance" button to automatically improve your prompt with more detail and creative elements
4. **Customize settings**:
   * Select the **number of images** to generate (1-4 images)
   * Choose your preferred **aspect ratio** from the available options
5. **Generate** - Click the generate button to create your images

<Frame>
  <video src="https://storage.googleapis.com/remade-v2/website/text-to-image.mp4" autoPlay loop muted playsInline style={{ width: "100%", height: "auto" }} />
</Frame>

## Selecting an Image Model

Choose from multiple AI models, each optimized for different styles and purposes:

| Model            | Best For                      | Strengths                                       |
| ---------------- | ----------------------------- | ----------------------------------------------- |
| **Runway Gen 4** | Photorealistic images         | High-quality realistic photos, natural lighting |
| **Recraft V3**   | Artistic and stylized content | Creative interpretations, artistic styles       |
| **Flux Kontext** | Consistent Characters         | High prompt adherence,                          |
| **GPT Image**    | General purpose               | Balanced results, good prompt adherence         |
| **Flux**         | Versatile generations         | Wide range of styles, reliable output           |
| **Imagen 4**     | High-resolution images        | Sharp detail, excellent text rendering          |
| **Ideogram**     | Text-heavy designs            | Perfect for images with text elements           |

<Tip>
  Try different models with the same prompt to see which style works best for
  your project. Some models support additional features like reference images.
</Tip>

## Adding Images as Reference

Reference images help guide your AI generation by providing visual context and style direction.

### How to Add References

1. **Select an existing image** - Click on any image in your canvas
2. **Set as reference** - Click the "Reference" button that appears
3. **Generate with reference** - Write your prompt and generate as normal

<Frame>
  <video src="https://storage.googleapis.com/remade-v2/website/references.mp4" autoPlay loop muted playsInline style={{ width: "100%", height: "auto" }} alt="Demonstrating how to add and use image references" />
</Frame>

<Note>
  Not all models support reference images. Check the model capabilities when
  selecting your AI model for generation.
</Note>

### Using References Effectively

When using reference images, you can:

* **Mention specific elements** - Reference particular aspects like "use the lighting from the reference image" or "apply the color palette from the reference"
* **Style transfer** - Ask the AI to generate in the same artistic style as your reference
* **Composition guidance** - Use reference images to guide the overall layout and composition
* **Color schemes** - Extract and apply color palettes from reference images

## Tips for Better Results

* **Be descriptive** - Include details about style, mood, lighting, and composition
* **Use the enhance feature** - Let AI improve your prompts for more creative results
* **Experiment with models** - Different models excel at different types of content
* **Iterate and refine** - Generate multiple variations and build on the best results

***

## Next Steps

<Card title="Image Editing" icon="paintbrush" href="/canvas/image-editing">
  Learn how to edit, enhance, and modify your generated images with powerful
  AI-powered editing tools including inpainting, outpainting, upscaling, and
  more.
</Card>