Multiple Image Inputs
Use multiple reference images at once to guide the AI. Arrange your references on the canvas, place input frames over each one, and generate images that draw from all of them simultaneously.
Add reference images
Upload or generate the images you want to use as references. Drag them onto the canvas and arrange them however you like, think of it as building a visual mood board.
Add input frames
Switch to Edit mode and click the multiple inputs button. Add extra input frames and position each one over a different reference image. Each frame captures what's inside it as a separate visual input for the AI.
Write your prompt and generate
Describe what you want the AI to create based on all the references. The AI sees every input frame as context and blends the visual information together to generate the result.
Up to 6 inputs
Use the main generation frame plus up to 5 additional input frames. Each one captures a different part of the canvas as a reference.
Visual mood boards
Arrange reference images on the canvas like a mood board. The AI uses all of them to understand what you're going for.
Style and composition guidance
Use one image for style, another for composition, a third for color palette, the AI blends them all.
Works with any source
Mix uploaded photos, AI-generated images, and hand-drawn elements. Everything on the canvas can be an input.
Why it matters
Show the AI what you mean instead of describing it, combine visual references for more precise results.
Use cases
- Style transfer from reference images
- Mood board to final image
- Combining product photos with background references
- Architectural concept from multiple source photos