Prompt templates for photos, illustrations, characters and logos — tap to copy, then describe your subject.
Tap Copy on any prompt, paste it into your AI assistant, and replace the [bracketed] parts with your own details.
Photo-realistic
Product shot
[product], professional product photography, on a [surface or background], soft studio lighting, shallow depth of field, high detail, photorealistic, 4k
Portrait
portrait of [subject], [lighting — e.g. natural window light], [mood], shot on an 85mm f/1.4 lens, shallow depth of field, high detail, photorealistic
Scene / landscape
[scene description], [time of day], [weather or atmosphere], cinematic lighting, wide angle, highly detailed, photorealistic
Illustration & art
Flat vector illustration
flat vector illustration of [subject], simple shapes, [colour palette], minimal, clean, modern, white background
In a specific style
[subject] in the style of [art style — e.g. watercolour / art deco / cyberpunk / hand-drawn], [colour mood], detailed
Character design
character design of [character description], full body, [art style], [colour palette], concept art, character sheet
Logos & icons
Logo concept
minimalist logo concept for [brand or idea], [style — e.g. geometric / lettermark], [colours], simple, flat, vector, on a white background
AI logos are a great starting point for ideas, but redraw the final version in a vector tool for real use.
Simple icon
simple line icon of [object], single colour, minimal, consistent stroke width, on a transparent background
No prompts match your search.
These templates work with Midjourney, DALL-E, Stable Diffusion, Leonardo and similar tools. Each generator has its own quirks, so treat them as starting points and adjust.
How image prompts are structured
Image prompts work differently from chat prompts — they're less a sentence and more a stack of descriptors: subject, then style, then lighting, then quality and detail. The order matters less than the coverage; the more of those layers you specify, the more control you get. Start with the templates here, then swap in your subject and experiment with the style and lighting terms.
Terms worth knowing
A few keywords reliably change results. For realism, naming a camera or lens ("85mm", "shot on film") and lighting ("soft studio lighting", "golden hour") pushes toward photography. For art, naming a medium or style ("watercolour", "flat vector", "concept art") sets the look. Quality tags like "highly detailed" and "4k" sharpen the output. And most tools support aspect ratio settings (square, portrait, wide) — check your generator's syntax for that, since it differs between them.
FAQ
Which image generators do these work with?
They're written as plain descriptive templates, so they work with Midjourney, DALL-E, Stable Diffusion, Leonardo and most others. Some tools have extra syntax (like aspect-ratio flags) you can add on top.
Can I use AI-generated images commercially?
It depends on the tool's licence and your local laws, which are still evolving. Check the specific generator's terms before using an image commercially, and be cautious with logos and anything resembling existing brands or artists.