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

# Generate an Image

> Create images from text prompts. Choose your aspect ratio and describe what you want.

Describe what you want and the agent generates one or more images on the canvas. You can generate from the home dashboard, from the chat, or directly on the canvas.

## How to generate

<Steps>
  <Step title="Select Image as the artifact type">
    On the home dashboard, click **Image** from the artifact type shortcuts. Or type your prompt directly in any Agent Mode chat.
  </Step>

  <Step title="Choose an aspect ratio">
    Select an aspect ratio before sending: 21:9, 16:9, 4:3, 1:1, 3:4, 9:16, or 9:21.
  </Step>

  <Step title="Describe what you want">
    Type a detailed description of the image. Include the subject, setting, lighting, mood, colors, and style.
  </Step>

  <Step title="The image appears on the canvas">
    The agent generates the image and places it on the canvas. You can generate multiple images at once by asking for a specific number.
  </Step>
</Steps>

You can also generate images directly on the canvas by clicking the **Generate Image** button in the canvas toolbar.

## Example prompts

<Prompt description="Product shot for a bakery's Instagram">
  Generate a photorealistic top-down flat lay of a rustic wooden table with 3 fresh croissants, a small jar of strawberry jam, a cup of black coffee in a white ceramic mug, and a linen napkin. Warm morning light from the left. Slightly out of focus green plant in the top corner. Clean, minimal, not cluttered. Aspect ratio 1:1 for Instagram feed.
</Prompt>

<Prompt description="Logo mockup for a fitness studio">
  Generate a clean mockup of a fitness studio storefront with a matte black sign above the door that reads "CoreFit Studio" in a modern sans-serif font. The sign is backlit with a subtle lime green (#84CC16) glow. Glass windows showing gym equipment inside, slightly blurred. Evening lighting with warm street lights. Aspect ratio 16:9.
</Prompt>

<Prompt description="Menu design for a restaurant">
  Generate a menu design for The Green Table restaurant. Single page, portrait orientation. Header with the restaurant name in an elegant serif font, dark green (#2D5016) text on a cream (#FFF8F0) background. Three sections: Starters (3 items), Mains (4 items), Desserts (2 items). Each item has a name in bold, a one-line description, and a price. Use a thin gold line between sections. Small herb illustration in the bottom corner. Aspect ratio 3:4.
</Prompt>

<Prompt description="Social media graphic for a marketing agency">
  Generate a professional social media graphic for BrightPath Marketing announcing "Q2 Results Are In". Dark navy (#1B2A4A) background with a large bold white headline. Below the headline, 3 stat callouts in teal (#00b6b6) rounded cards: "340 Leads", "\$52 CPL", "95 New Patients". Small agency logo placeholder in the bottom right corner. Clean, data-driven design. Aspect ratio 1:1.
</Prompt>

## Generate multiple images

Ask the agent to generate more than one image at once. They generate in parallel and appear on the canvas side by side.

<Prompt description="Generate multiple product shots at once">
  Generate 4 product photos for my bakery: a sourdough loaf on a cutting board, a chocolate croissant on a plate, a selection of 6 cookies arranged in a circle, and a loaf of banana bread sliced and styled. All on a light wood table with warm natural lighting. 1:1 aspect ratio.
</Prompt>

## What you cannot do

* You cannot control exact pixel dimensions. Only aspect ratio.
* You cannot generate images larger than 4K resolution (depends on the model).
* You cannot generate NSFW content (safety filters are active).
* You cannot download images in vector format (SVG). All output is raster (PNG).

## Next steps

<CardGroup cols={2}>
  <Card title="Remove Background" icon="image" href="/remove-background">
    Strip the background from your generated image.
  </Card>

  <Card title="Edit Text in Images" icon="font" href="/edit-text-in-images">
    Change text in an image without regenerating the whole thing.
  </Card>
</CardGroup>