Output image dimensions.
Number of variations to generate (1-4).
High fidelity preserves more details from input images (faces, logos, textures). Uses more tokens but provides better quality.
Scan this QR code with your mobile device to upload an image