How to Use Leonardo AI: Create Stunning AI Art & Images (2026)

Leonardo AI is a creative AI platform that makes professional image generation accessible to everyone. With its intuitive interface, pre-trained models, and powerful features like real-time canvas and motion generation, Leonardo has quickly become a favorite among designers, game developers, and content creators.

This tutorial covers everything you need to start creating with Leonardo AI, from basic image generation to advanced techniques like model fine-tuning and real-time editing.

TL;DR — Quick Start

1. Sign up free at leonardo.ai — 150 tokens/day on free plan
2. Choose a model: Leonardo Phoenix for general use, PhotoReal for photos
3. Write a prompt and click Generate
4. Refine results using the Canvas editor for precise edits

Leonardo AI Pricing

Plan	Price	Daily Tokens	Features
Free	$0	150	Basic generation, community models
Apprentice	$12/mo	8,500/mo	Faster generation, more models
Artisan	$30/mo	25,000/mo	Priority queue, all features
Maestro	$60/mo	60,000/mo	Maximum speed, API access

Token costs vary by feature:

• Standard image: 1-2 tokens
• High-resolution: 4-8 tokens
• Real-time canvas: 1 token per edit
• Motion generation: 25 tokens per video

Getting Started

Step 1: Create an Account

Visit leonardo.ai and sign up with Google, Apple, or email. No credit card required for the free plan.

Step 2: Navigate the Dashboard

The main dashboard shows:

• Home Feed: Community creations for inspiration
• Image Generation: Your main workspace
• Canvas Editor: Real-time editing tool
• Training & Datasets: Custom model fine-tuning
• Motion: Animate static images

Step 3: Generate Your First Image

1. Click Image Generation in the sidebar
2. Select a model (start with Leonardo Phoenix)
3. Type your prompt in the text box
4. Adjust settings (or keep defaults)
5. Click Generate

Choosing the Right Model

Leonardo offers several specialized models:

Model	Best For	Style
Leonardo Phoenix	General purpose	Versatile, high quality
Leonardo Diffusion XL	Artistic images	Creative, painterly
PhotoReal v2	Photographs	Photorealistic
DreamShaper v7	Fantasy/Concept art	Dreamy, imaginative
Leonardo Vision XL	Design assets	Clean, commercial
Anime Pastel Dream	Anime/Manga	Japanese animation style

Tip: Start with Leonardo Phoenix — it handles most prompts well and produces the highest quality output for general use.

Writing Effective Prompts

Basic Structure

[Subject] in [setting], [style/medium], [lighting], [mood], [quality tags]

Example:

A cozy bookshop interior at sunset, warm lighting through dusty windows, photorealistic, ultra detailed, soft golden hour glow, 8K

Prompt Enhancement

Leonardo has a built-in Prompt Magic feature that automatically improves your prompts. Toggle it on for:

• Added detail descriptors
• Improved composition suggestions
• Style-specific enhancements

Negative Prompts

Click “Add Negative Prompt” to exclude unwanted elements:

blurry, low quality, watermark, text, deformed, ugly, oversaturated

Style Presets

Leonardo offers one-click style presets:

• Cinematic: Movie-like lighting and composition
• Creative: Bold, artistic interpretation
• Dynamic: High-energy, action-oriented
• Environment: Landscape and scene focus
• Illustration: Clean, graphic style
• Photography: Realistic camera aesthetics
• Raytraced: 3D-rendered quality

Generation Settings

Image Dimensions

Choose from presets or set custom dimensions:

• 1:1 (1024×1024): Social media, profile images
• 16:9 (1360×768): Presentations, banners
• 9:16 (768×1360): Stories, mobile wallpapers
• 3:2 (1152×768): Photography standard
• 2:3 (768×1152): Portraits

Guidance Scale

• Range: 1-20
• Low (1-5): More creative freedom
• Medium (7-10): Balanced (recommended)
• High (12-20): Strict prompt adherence

Number of Images

Generate 1-8 images per prompt. Generating 4 images gives you variety while conserving tokens.

The Canvas Editor

Leonardo’s Canvas is a real-time editing workspace that lets you modify images with AI assistance.

Key Canvas Features

Inpainting: Paint over areas to regenerate specific parts

1. Upload or drag a generated image onto the Canvas
2. Use the brush tool to mask the area you want to change
3. Type what should replace it
4. Click Generate to fill the masked area

Outpainting: Extend images beyond their original borders

1. Place your image on the canvas
2. Move it to create empty space on any side
3. Describe what should fill the empty area
4. Generate to seamlessly extend the image

Sketch to Image: Draw rough shapes and convert to finished art

1. Use drawing tools to sketch basic shapes
2. Add a text prompt describing the desired result
3. The AI interprets your sketch and generates a polished version

Canvas Tips

• Use lower guidance scale (5-7) for more natural blending
• Generate multiple variations and pick the best
• Work in layers — fix one area at a time
• Match the prompt style to the original image

Advanced Features

Image to Image

Transform existing images with AI:

1. Upload a reference image
2. Write a prompt describing desired changes
3. Adjust Init Strength (0.1-1.0):

• Low (0.1-0.3): Heavy changes, loose reference
• Medium (0.4-0.6): Balanced transformation
• High (0.7-0.9): Minor changes, close to original

AI Upscaling

Increase image resolution without losing quality:

• Select any generated image
• Click “Upscale” for 2x or 4x enlargement
• Universal Upscaler preserves details while adding sharpness
• Costs additional tokens based on upscale factor

Motion Generation

Turn static images into short video clips:

1. Select a generated image
2. Click “Generate Motion”
3. Choose motion strength (1-10)
4. Select motion type: camera pan, zoom, or AI-interpreted motion
5. 4-second clips are generated

Fine-Tuning Custom Models

Train Leonardo on your own images:

1. Go to Training & Datasets
2. Create a new dataset
3. Upload 10-30 images of your subject
4. Set training parameters (category, resolution)
5. Train (takes 30-60 minutes)
6. Use your custom model for generation

Best practices for training data:

• Use high-quality, consistent images
• 15-20 images is the sweet spot
• Include variety in angles and lighting
• Keep the subject consistent across images

Practical Use Cases

Social Media Content

• Generate eye-catching post images
• Create consistent brand imagery
• Make story backgrounds and cover photos

Game Development

• Concept art for characters and environments
• UI elements and icons
• Texture generation for 3D assets

Marketing Materials

• Product mockups and lifestyle shots
• Ad creative variations
• Blog post header images

Personal Projects

• Custom wallpapers and art prints
• Avatar and profile picture creation
• Gift card and invitation designs

Tips for Best Results

1. Start broad, then refine: Generate with a simple prompt first, then add details in subsequent generations
2. Use image2image: Take your best generation and use it as a starting point for refinements
3. Leverage community models: Browse community-created models for specialized styles
4. Save your best prompts: Build a prompt library for consistent results
5. Experiment with models: Different models excel at different subjects
6. Use Canvas for fixes: Don’t regenerate the whole image — fix specific areas with Canvas

FAQ

Is Leonardo AI free?

Yes, the free plan provides 150 tokens per day, enough to generate roughly 75-150 standard images. Paid plans start at $12/month for more tokens and faster generation times.

How does Leonardo compare to Midjourney?

Leonardo offers a more accessible interface with a free tier, built-in editing tools (Canvas), and custom model training. Midjourney generally produces higher artistic quality for certain styles but requires Discord and has no free plan. Leonardo is better for iterative editing and commercial workflows.

Can I use Leonardo AI images commercially?

Yes, paid plan subscribers can use generated images for commercial purposes. Free plan users should check the current terms of service, as commercial use may be restricted. Always review the latest usage policy.

How do I get consistent characters?

Use the same detailed character description across prompts, including specific features like “woman with short red hair, green eyes, freckles, wearing a blue jacket.” For best consistency, fine-tune a custom model with reference images of your character.

What image formats does Leonardo export?

Leonardo exports images as PNG and JPEG. You can also download the upscaled versions. Canvas edits are saved in your workspace and can be exported at any time. For more insights, check out our guide on Best AI Image Generators. For more insights, check out our guide on Les meilleurs générateurs d’images IA en 2026. For more insights, check out our guide on Best ChatGPT Alternatives.

Conclusion

Leonardo AI strikes an excellent balance between power and accessibility. The free tier lets you experiment extensively, while the Canvas editor and model training features give professionals the tools they need for production work.

Start by generating images with Leonardo Phoenix, experiment with different prompts and settings, and explore the Canvas editor for fine-tuned control. As your needs grow, consider fine-tuning a custom model for consistent, brand-specific imagery.