How to create consistent character images with one reference photo
The Rundown: Create unlimited consistent character variations from a single reference image using the new Ideogram character consistency model.
🧰 Who is this useful for:
- Content creators building personal brands across platforms
- Digital marketers creating cohesive visual campaigns
- Game developers designing character concepts
- Authors visualizing book characters in different scenarios
STEP 1: Access the Character Feature
Head over to Ideogram and look for the “Character” icon in the left sidebar, it's represented by a face symbol. Click on this icon to access the character consistency feature.
Once inside, you'll see three options for creating your character: take a photo with your camera, upload an image from your device, or choose from your existing creations on the platform.
STEP 2: Upload and Prepare Your Reference Image
Select your reference photo by clicking “Upload” and choosing a high-quality image. For best results, use a photo with:
- Solid, uncluttered background
- Front-facing pose
- Clear, well-lit facial features
- High resolution
After uploading, the AI will automatically mask your face, isolating it from the background. If the automatic masking missed any areas you want included (like your neck or specific facial features), use the brush tool to manually edit the mask by erasing red areas to reveal more of the face you want preserved.
STEP 3: Configure Your Generation Settings
With your character reference ready, describe what you want to see in the text prompt field. If you need inspiration, click “Templates” to browse pre-made prompts like “statue portrait,” “sky portrait,” or “adventure scene.”
Next, customize your settings:
- Aspect Ratio: Choose from standard ratios or create a custom one (9:16 works great for social media)
- Model: Select the latest version (3.0) for character consistency
- Number of Images: Choose 2-4 images per generation
- Render Speed: Select "Quality" for the best results over "Turbo" or "Default"
- Magic Prompt: Toggle on if you want AI assistance improving your prompt, or off if your prompt is already detailed
STEP 4: Generate and Enhance Your Images
Click “Generate” and watch as the AI creates multiple variations of your character in different poses, expressions, and scenarios while maintaining facial consistency. The generation process typically takes 10-20 seconds even with quality settings.
Once generated, review your options and select your favorite image. For even higher quality, click “Upscale” on your chosen image to enhance the resolution and detail further. Keep in mind that upscaling might slightly alter facial features, so compare the results to ensure you're satisfied with the consistency.
Pro tip: Experiment with different prompts and scenarios to build a comprehensive library of your character in various situations.