Image prompts for Gemini
Gemini creates and edits images from the chat. Like ChatGPT, it understands full sentences well, and it's good at refining an image step by step.
With VisualCodes you start from a complete prompt instead of a loose idea: subject, lighting and camera style, colour and format, all in one sentence.
How to use it with Gemini
- 1Build the combination in the Builder.
- 2Tap “Try in Gemini”: the prompt is copied (with “Create an image:” in front) and Gemini opens.
- 3Paste it into the chat. For adjustments, ask for specific changes to the generated image.
Ready examples for Gemini
A warm, close social post. · Codes: /AmbientGlow /BokehPortrait /CozyMood
Create an image: Steaming cup of coffee on a table by the window, soft ambient glow, strong background bokeh, cozy intimate mood, 4:5, no text
Frozen action against the light, high contrast. · Codes: /CityStreet /FreezeAction /Backlight /HighContrastColor
Create an image: Skateboarder jumping in a city street, city street setting, urban buildings, pavement, traffic cues and everyday street atmosphere, frozen high-speed action, strong backlight, high colour contrast, 16:9, no text
Drink product with a single colour accent. · Codes: /IceCold /RedAccent /Smoke
Create an image: Soda bottle with droplets on a dark background, ice-cold drink look, single red accent, soft smoke atmosphere, 1:1, no text, no logo
Tips for Gemini
- Gemini doesn't take the prompt from the URL: that's why it's copied to the clipboard when you tap the button.
- Ask for one change at a time (“now with sunset light”): small changes work better than rewriting everything.
- State the format in words (“vertical 9:16 for a story”) so it's respected.
Build your own combination and tap “Try in Gemini” to take it straight there.
Open the Builder