Back to Guides
AI Design ToolsIntermediate
20 min

Creating Hero Images with Midjourney

Generate stunning hero images and visual assets for your UI designs using Midjourney's advanced AI.

Hero images make or break a landing page's first impression. Midjourney has become the go-to tool for designers who need unique, high-quality visuals without stock photo limitations or expensive photo shoots. This comprehensive guide will teach you how to create professional hero images that perfectly complement your UI designs.

1. Set Up Your Midjourney Account and Understand Parameters

Before creating hero images, you need to understand Midjourney's setup and key parameters.

Getting Started

Join the Midjourney Discord server and subscribe to a plan. The Basic plan ($10/month) is sufficient for most designers, but Standard ($30/month) offers unlimited relaxed generations.

Key Parameters to Know

--ar (aspect ratio): Use --ar 16:9 for standard hero images, --ar 21:9 for ultra-wide designs. --v 6 (version): Use the latest version for best quality. --s (stylize): Range 0-1000, higher values = more artistic, lower = more literal. --q (quality): 0.25, 0.5, 1, or 2. Use 1 for most hero images.

Understanding Prompt Weight

Use :: to separate prompt parts and add weights. For example: 'modern office::2 natural lighting::1' emphasizes the office twice as much as the lighting.

2. Craft Prompts Specifically for Hero Images

Hero images have specific requirements different from general AI art. Here's how to prompt for them:

Focus on Negative Space

Include 'with copy space' or 'left side clear for text' in your prompts. Example: 'modern workspace, natural lighting, minimalist desk setup, right third composition with left side clear for headline text --ar 16:9'

Specify Lighting and Mood

Lighting dramatically affects mood. Use: 'golden hour lighting,' 'soft natural light,' 'dramatic side lighting,' 'bright and airy,' or 'moody and atmospheric.'

Consider Brand Alignment

Include brand-relevant elements: 'tech startup aesthetic,' 'corporate professional,' 'creative agency vibe,' 'wellness brand feeling,' 'eco-friendly sustainable.'

Avoid Complexity

Hero images should be clean and not compete with UI elements. Use 'minimal composition,' 'clean background,' 'uncluttered scene,' 'simple elegant.'

3. Use Aspect Ratios and Style References Effectively

Technical control over output format and style is crucial for hero images that work in UI contexts.

Aspect Ratio Strategy

--ar 16:9 for standard desktop hero sections, --ar 21:9 for ultra-wide hero banners, --ar 3:2 for flexible responsive designs, --ar 4:5 for mobile-first hero images.

Style References (--sref)

Upload reference images to guide the aesthetic. Use existing brand photos or designs you like: '/imagine prompt: [your prompt] --sref [image URL]'

Character References (--cref)

Maintain consistent people or products across multiple generations using character references. Perfect for brand mascots or recurring visual elements.

Chaos Parameter

Use --chaos 0-10 to control variety. Lower chaos (0-20) for consistent brand imagery, higher chaos (50-100) for exploring creative options.

4. Upscale and Refine Your Generated Images

Initial generations are starting points. Here's how to polish them for production use:

Selective Upscaling

Generate 4 variations (default), then upscale only the best ones. Use 'Upscale (2x)' or 'Upscale (4x)' for maximum quality hero images.

Variation and Remix

Found an image that's 90% there? Use 'Vary (Region)' to regenerate specific parts while keeping the rest. Perfect for adjusting problematic areas.

Subtle and Strong Variations

'Vary (Subtle)' for small tweaks preserving the original. 'Vary (Strong)' for significant changes while maintaining the core composition.

Zoom Out for Context

If your hero image feels cropped, use 'Zoom Out 1.5x' or 'Zoom Out 2x' to expand the canvas while maintaining the subject.

5. Integrate AI Images into Your Design Workflow

Getting great images is only half the battle. Here's how to use them effectively in UI designs:

Post-Processing in Figma

Import into Figma and add gradient overlays for better text readability. Use blend modes (multiply, overlay) to integrate with your color scheme. Apply subtle blur or noise for texture.

Responsive Considerations

Generate multiple aspect ratios of the same concept for different breakpoints. Ensure important elements aren't cropped on mobile. Test how the image works at various viewport sizes.

Accessibility

Ensure sufficient contrast between hero image and text overlays (4.5:1 minimum). Provide descriptive alt text for all hero images. Consider reduced motion alternatives for animated hero sections.

File Optimization

Export as WebP for best compression and quality balance. Use progressive JPEGs for faster perceived loading. Implement lazy loading for below-the-fold hero images.

Key Takeaways

  • Use aspect ratios that match your target viewport: --ar 16:9 for desktop, --ar 4:5 for mobile
  • Always specify negative space in prompts for text overlay areas
  • Leverage style references (--sref) to maintain brand consistency
  • Upscale selectively and use variation tools to refine results
  • Post-process in design tools with overlays and blend modes for UI integration
  • Generate multiple versions for responsive designs and A/B testing

Ready to learn more?

Explore more guides to master AI-powered design

View All Guides