Image Generation

Generate images directly in chat with Gemini or xAI models.

Image Generation

Generate images directly in your conversation using Gemini or xAI models. Describe what you want, and the AI creates it.

Supported Models

ProviderModelNotes
Googlegemini-2.5-flash-imageFast image generation
Googlegemini-3-pro-image-previewPreview model
xAIgrok-2-image-1212xAI image generation
xAIgrok-imagine-imageStandard quality
xAIgrok-imagine-image-proHigher quality

Quick Start

1. Select an Image Model

  1. Click the provider selector in the toolbar
  2. Choose Gemini or xAI
  3. Select an image generation model

2. Choose Aspect Ratio

Click the aspect ratio selector in the toolbar:

RatioBest For
1:1Profile pictures, icons
16:9Banners, presentations
9:16Stories, mobile wallpapers
4:3Standard photos
3:4Portraits
3:2Photography
2:3Portrait photography
4:5Instagram posts
5:4Landscape orientation
21:9Ultrawide, cinematic

Note: Gemini doesnโ€™t support โ€œautoโ€ aspect ratio. Select a specific ratio. xAI offers โ€œautoโ€ which lets the model choose.

3. Describe Your Image

Type a description and send:

You: Generate a minimalist logo for a coffee shop called "Morning Brew"

The image appears in the response.

Tips for Better Images

Be Specific

โŒ "A cat"
โœ… "A fluffy orange tabby cat sitting on a windowsill, soft morning light, photorealistic"

Include Style

โœ… "in the style of Studio Ghibli"
โœ… "minimalist line art"
โœ… "photorealistic"
โœ… "watercolor painting"

Specify Details

  • Subject: Whatโ€™s in the image
  • Setting: Background, environment
  • Mood: Bright, dark, dramatic, calm
  • Style: Art style, medium
  • Colors: Color palette

Example Prompts

Logo Design

Minimalist logo for a tech startup, abstract geometric shape,
blue and white colors, clean modern design, vector style

Illustration

Cozy coffee shop interior, warm lighting, wooden furniture,
plants by the window, watercolor illustration style

Photo-Style

Mountain landscape at sunset, golden hour lighting,
dramatic clouds, photorealistic, wide angle

Character

Cute robot character, round design, friendly expression,
holding a coffee cup, Pixar animation style

Viewing Generated Images

In Chat

Generated images appear inline in the conversation.

Click an image to open the lightbox:

  • Full-size view
  • Navigate between images
  • Download
  • Copy to clipboard
  • Jump to source conversation

All generated images are saved to the Gallery:

  • Browse all images
  • Favorite best ones
  • View by conversation

Using Images

Copy to Clipboard

  1. Click the image to open lightbox
  2. Click the copy button
  3. Paste anywhere

Download

  1. Click the image
  2. Click the download button
  3. Save to your computer

Share

Copy the image and share in:

  • Documents
  • Presentations
  • Social media
  • Messages

Limitations

Content Policy

Both providers have content policies:

  • No explicit content
  • No copyrighted characters
  • No harmful imagery

If a prompt violates policy, youโ€™ll see an error.

Quality Varies

  • Simple prompts โ†’ unpredictable results
  • Detailed prompts โ†’ more consistent
  • Multiple attempts may be needed

Rate Limits

Check your providerโ€™s documentation for current rate limits.

Troubleshooting

โ€œImage generation failedโ€

  • Try a simpler prompt
  • Check content policy compliance
  • Verify API key has access

Images look wrong

  • Add more detail to prompt
  • Specify style explicitly
  • Try different aspect ratio

Slow generation

  • Image generation takes 10-30 seconds
  • Complex prompts take longer
  • Check network connection

Model not available

  • Ensure youโ€™re using the correct model
  • Check provider status
  • Verify API key permissions

Generated images automatically appear in the Gallery:

  1. Open Gallery from header
  2. View all images including generated ones
  3. Favorite, copy, or download

โ†’ Gallery guide

  • Gallery View โ€” Browse all images
  • Gemini โ€” Google provider setup
  • xAI โ€” xAI provider setup