What makes a good AI image prompt?

A clear subject, what it is doing, where it is, the visual style, lighting, camera angle and mood. Specific descriptions ('golden-hour light on a Jaipur rooftop') beat vague ones ('nice picture').

Does this tool generate images?

No. It writes the prompt text for you, which you then paste into the image generator of your choice. Nothing is sent to any AI service from this page.

What is a negative prompt?

Some generators, such as many Stable Diffusion interfaces, let you list things you do not want, like 'blurry, extra fingers, text'. Other tools ignore it, so you can simply leave it out.

AI Tools / AI Image Prompt Builder

🎨 AI Image Prompt Builder

Fill in what you can and the prompt updates as you type. Copy it into Midjourney, DALL·E, Stable Diffusion, Gemini or ChatGPT.

Subject (who or what?) Action / pose Setting Style Lighting Camera / composition Mood Colours Aspect ratio Format for Extra details (optional)

The anatomy of a great image prompt

Image generators do best when you describe a picture the way a photographer or art director would. A strong prompt usually covers these parts, roughly in this order:

Subject: who or what, with one or two specific details ("an elderly tea seller with a warm smile").

Action: what they're doing ("pouring chai from a height").

Setting: where and when ("at a railway platform at dawn").

Style: photo, painting, 3D, illustration, a folk-art style…

Lighting: often the single biggest quality boost.

Camera and composition: close-up, wide-angle, aerial, depth of field.

Mood and colours: the feeling you want the image to have.

Before and after

Weak prompt	Strong prompt
"a cat"	"A ginger cat curled up on a windowsill during monsoon rain, soft overcast light, shallow depth of field, cosy and warm mood, photorealistic photograph"
"Indian wedding"	"A joyful Indian wedding baraat dancing in the street at night, marigold garlands, string lights, motion blur, cinematic film still, vibrant colours"

Weak prompt

Strong prompt

"a cat"

"A ginger cat curled up on a windowsill during monsoon rain, soft overcast light, shallow depth of field, cosy and warm mood, photorealistic photograph"

"Indian wedding"

"A joyful Indian wedding baraat dancing in the street at night, marigold garlands, string lights, motion blur, cinematic film still, vibrant colours"

Tips by tool

ChatGPT, DALL·E and Gemini understand natural sentences well. Describe the scene clearly, and you can ask for edits afterwards in plain language.

Midjourney responds well to concise, descriptive phrases. Parameters such as --ar 16:9 set the aspect ratio.

Stable Diffusion interfaces often work best with comma-separated keywords, and many support a negative prompt listing what to avoid.

Text in images is still hit-and-miss for many generators. Keep any wording short and put it in quotes.

Use AI images responsibly

Don't create realistic images of real people without their consent, or images meant to mislead.

Check each tool's terms for commercial use before using images in ads or products.

Label AI-generated images where your platform or audience expects it.