Step 1: Open Realtime Canvas and Configure Style

Launch Leonardo AI and select the “Realtime Canvas” tab. Choose your target model — this determines the visual style applied to your sketches. If you’ve trained a custom Phoenix model, select it for brand-consistent output. Otherwise, pick a Leonardo model that matches your project’s aesthetic (e.g., “Anime” for character concepts, “3D Render” for game environments, “Sketch” for rough ideation). Set the creativity slider — lower values produce outputs closer to your sketch strokes; higher values give the AI more interpretive freedom.

Step 2: Set Up Your Canvas Workspace

Configure the canvas size and background. For character concepts, use a portrait-oriented canvas (800×1000px) with a neutral gray background. For environments, use landscape orientation (1200×800px) with a dark background that won’t interfere with generated lighting. Enable the “Instant Generation” toggle so images update as you draw rather than on-demand. This real-time feedback is the core of the workflow — you see results immediately, allowing continuous creative steering.

Step 3: Sketch and Steer in Real Time

Begin sketching your concept with rough strokes. Draw a character silhouette, an architectural outline, or a landscape horizon. As your strokes appear, the AI immediately transforms them into a styled rendering. Watch the output evolve as you add more detail. This is where the magic happens — you’re not waiting for generations, you’re painting with AI as your live collaborator. If the AI interprets your strokes differently than intended, adjust your drawing or modify the creativity slider to give more or less interpretive freedom.

Step 4: Refine with Selective Generation

When you’ve established a promising direction, switch from continuous real-time mode to selective generation. Highlight specific areas of your sketch (e.g., just the character’s face, just the building’s entrance) and generate refinements for those regions only. This selective approach lets you iterate on specific elements without disrupting the overall composition — far more efficient than regenerating the entire image each time you want to adjust one detail.

Step 5: Export and Continue in Your Pipeline

Once you’re satisfied with a concept direction, export the Realtime Canvas output as a high-resolution image. Import it into Photoshop, Procreate, or your preferred painting tool for manual refinement — adding details the AI missed, correcting anatomy, enhancing lighting. The Realtime Canvas output serves as a strong foundation that eliminates hours of initial sketching time, letting you focus your manual effort on the finishing touches that define professional concept art quality.

Step 1: Prepare Your Reference Dataset

Gather 10-50 high-quality images that exemplify your target visual style. For a brand, this might include existing marketing photography, product shots, and design mockups. For a game, collect concept art, character designs, and environment paintings. Quality matters more than quantity — every reference image should perfectly represent the style you want to replicate. Remove any outliers that deviate from your core aesthetic, as they’ll confuse the model.

Step 2: Create a New Phoenix Dataset

In Leonardo AI, navigate to the “Training” section and click “Create New Dataset.” Upload your reference images and assign descriptive tags to each one — not generic tags like “image” but specific ones like “dark fantasy castle, moody lighting, stone textures, atmospheric fog.” These tags teach the model what elements matter in your style. Tag quality directly impacts model quality, so be thorough and precise.

Step 3: Configure and Launch Training

Select your dataset and configure training parameters. Set the training strength (higher = closer to your reference style, lower = more creative freedom). Choose whether to enable style-only training (preserves composition flexibility) or full training (replicates both style and composition patterns). Click “Start Training” — the process typically takes 15-30 minutes. Leonardo will display progress updates and a quality score upon completion.

Step 4: Test Your Custom Model

Once training completes, your custom model appears in the model selector dropdown. Switch to it and generate several test images using prompts that describe scenes different from your reference set but in your trained style. For example, if your references were fantasy castles, prompt for a fantasy marketplace or a fantasy ship — different content, same style. Evaluate consistency across generations: do they all share the same color palette, lighting approach, texture detail, and overall mood? If results are inconsistent, add more reference images and retrain.

Step 5: Build Your Asset Library

With a validated custom model, start generating your full asset library. Create a prompt template system: “In [your trained model style], [scene description], [lighting], [composition].” Use this template for every generation to maintain maximum consistency. Save all outputs to your Leonardo project gallery, organized by category (characters, environments, items, UI elements). Export at your required resolution and format — Leonardo supports up to 4K output with transparent backgrounds for game sprites.