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

# Image Generation

> Generate and edit images with AI

Your assistant can generate images from text descriptions and edit existing images using Google's Gemini image models.

## Available models

| Model                               | Strength              | Best for                             |
| ----------------------------------- | --------------------- | ------------------------------------ |
| **Gemini 3.1 Flash** (Nanobanana 2) | Fast, general-purpose | Quick drafts, iterative exploration  |
| **Gemini 3 Pro** (Nanobanana Pro)   | Highest quality       | Final assets, detailed illustrations |

## Generating images

Ask your assistant to create an image by describing what you want:

* *"Generate a photorealistic golden retriever sitting in a park"*
* *"Create a minimalist logo for a coffee shop called 'Brew'"*
* *"Draw a technical diagram of a microservices architecture"*

You can specify aspect ratio and detail level:

* *"Generate a 16:9 banner image of a mountain landscape at sunset"*
* *"Create a square icon for a mobile app"*

### Supported aspect ratios

`1:1` · `3:4` · `4:3` · `9:16` · `16:9` · `3:2` · `2:3` · `4:5` · `5:4` · `21:9`

## Editing images

You can also ask your assistant to modify existing images:

* *"Remove the background from this image"*
* *"Change the sky to a sunset in this photo"*
* *"Add a watermark to the bottom-right corner"*

Attach the image to your message or reference one the assistant previously generated.

## Credits

Image generation uses credits from your Coral plan, same as text model requests. The cost varies by model and image size.

<Tip>
  Start with the Flash model for quick iterations, then switch to Pro for the final version to conserve credits.
</Tip>