Text to ImageText to Image
Describe a subject, scene, and style to create a new image.
AI Image to Image
Upload a reference. Describe your changes.
Popular Models
Choose your image-to-image model
Pick a model for your creative goal and explore different results.
Start with a reference. Explore new directions.
Upload an image → describe your changes → choose a model → generate and download.
Start with a template
Choose a portrait, product, or poster template. Upload your image and adapt the prefilled prompt to your goal.
Specify what to keep and change
Keep a person, product, or composition while exploring backgrounds, materials, lighting, and styles. Describe the details you want to retain.
Combine multiple references
Upload separate references for a person, outfit, product, or scene and explain each image’s role. Reference limits depend on the selected model.
Create variations from one image
Try new backgrounds, styles, and aspect ratios with the same reference to create avatars, social covers, product images, or ad visuals.
Keep exploring PixZoo
Keep creating images and videos
Create images from text or turn an existing image into a video.
Frequently Asked Questions
Using reference images, prompts, and generation settings.
How is image-to-image different from text-to-image?
Text-to-image uses a written description. Image-to-image adds visual references, helping you build on an existing person, product, composition, or style.
How should I describe my changes?
First state what should stay the same, then describe your changes. For example: “Keep the person’s facial features and pose. Replace the background with light gray and use soft studio lighting.”
What can I create with image-to-image?
Try portrait restyling, product backgrounds, lighting adjustments, poster variations, and compositions using multiple references. Use the image editor for precise local adjustments.
How many reference images can I upload?
Check the selected model’s limits and supported formats in the generator. When using multiple references, describe each image’s role in your prompt.
How can I retain a person’s or product’s features?
Use clear, unobstructed references and specify the facial features, outfit, product shape, or brand colors to retain. Review the key details after generation and refine your prompt as needed.
Which models can I use?
Choose reference-capable models such as GPT Image 2, Nano Banana 2, Nano Banana Pro, and Seedream 5.0 in the generator. Available models and settings are shown there.
How many credits does generation use?
Credit usage depends on the model, output settings, and image count. The generator shows the estimated cost before submission. See the pricing page for plans and credit details.
Can I keep editing the result?
Download the result or use it as a reference for another generation. Use the image editor for local changes, or image-to-video to create motion.