Image prompt basics: style, camera, lighting, quality

Short answer: a strong image prompt states four layers separately — what to depict (subject), which visual style (realistic, anime, cyberpunk, etc.), what camera or composition (close-up, drone, wide angle), and what kind of lighting (golden hour, studio, neon).

Stating these separately instead of blending them into one vague sentence gives more consistent results across models like Midjourney, Stable Diffusion, and Flux. 'A nice coffee cup' is far less predictable than 'a coffee cup on a table, cinematic style, close-up shot, golden hour lighting, 4K quality.'

Model differences apply here too: Midjourney leans more artistic and interpretive, Stable Diffusion gives finer technical control (seed, sampler), and Ideogram is notably better at rendering text inside an image.

PromptMaker's image category exposes these four layers — style, camera, lighting, quality — as separate buttons, so you pick each one and the app combines them.