To help get the best results when using Google Gemini (Nano Banana) or GPT Image, consider structuring your prompt.
Define Core Elements
Subject Describe who or what appears (e.g., “a stoic robot barista with glowing blue optics”)
Composition Specify framing (e.g., extreme close-up, wide shot, low angle)
Action Define what’s happening (e.g., “brewing coffee,” “casting a spell”)
Location Set the scene (e.g., “futuristic cafe on Mars,” “sun-drenched meadow”)
Style Choose aesthetic (e.g., 3D animation, photorealistic, watercolor)
Editing Instructions Be direct for modifications (e.g., “change the tie to green”)
Advanced Controls:
Add camera details eg: low-angle shot, shallow depth of field f/1.8
Define lighting eg: golden hour backlighting, muted teal color grading
Specify exact text appearance eg: “‘URBAN EXPLORER’ in bold white sans-serif font”