AI Prompt Generator
Type a short idea and choose where you’ll use it. You get one detailed prompt, written the way that image or video model responds best.
5 of 5 free tries left today
Your prompt
Type your text and press Generate prompt.
Results usually arrive in a few seconds.
From a few words to a prompt that works
Image models do their best work with specific prompts, but most ideas start as a few words. This AI prompt generator bridges the gap. You give it the idea, it keeps your subject, and it writes the details a model needs: setting, light, camera, style and mood.
Each model reads prompts differently, so you choose the target first. Midjourney gets a line of descriptors, Stable Diffusion gets weighted tags with a negative prompt, and the video models get a shot with action and camera movement.
The five layers of a strong prompt
- 01
Subject
Who or what, doing what.
- 02
Setting
Where and when: place, time of day, weather.
- 03
Light
Direction, color and quality of the light.
- 04
Camera or medium
Lens and angle for photos; paint, ink or 3D for art.
- 05
Mood
The feeling the picture should leave.
The generator fills in the layers your idea leaves out. Anything you do specify, it keeps.
Written for the model you use
Image models: ChatGPT, Nano Banana (Gemini), Midjourney, Flux, Stable Diffusion.
Video models: Sora, Veo, Kling, Runway.
Use it as an AI art prompt generator for still images, or switch to a video model to get a short shot description instead.
Frequently asked questions
How does the AI prompt generator work?
You type a short idea, such as “a fox reading in a cozy library at night”, and choose the model you’ll use. The generator keeps your subject and adds the details that model responds to: setting, lighting, camera, style and mood.
Which models can I write prompts for?
ChatGPT, Nano Banana (Gemini), Midjourney, Flux and Stable Diffusion for images, and Sora, Veo, Kling and Runway for video. Each gets its own format: Midjourney a line of descriptors, Stable Diffusion tags plus a negative prompt, video models a shot with camera movement.
What makes a good AI art prompt?
A clear subject, a setting, light, a style or medium, and a mood, in that order of importance. Specific beats vague: “golden-hour side light” does more than “beautiful lighting”.
I already have a prompt. Can I improve it?
Yes: use the prompt enhancer. It rewrites an existing prompt instead of starting from an idea.
Is it free?
Yes, with no account. Everyone gets 5 free tries a day across all our tools. What you type isn’t saved by us.
More free tools
- Image to prompt generatorTurn any picture into a ready-to-use prompt for Midjourney, Flux, Stable Diffusion or ChatGPT.
- AI image describerGet a clear written description of any picture: a caption, a short summary or a detailed paragraph.
- Image to text converterExtract text from screenshots, photos and scans, line by line or as plain paragraphs, ready to copy.
- Image translatorTranslate the text in a photo or screenshot into English or nine other languages, as text you can copy.
- Alt text generatorUpload an image and get three alt text options under 125 characters, with an optional focus keyword and page context.
- Alt Text for SEOWrite SEO alt text with a focus keyword and page context. Free, no account.
- AI prompt enhancerPaste a rough prompt and get a sharper version for ChatGPT, Midjourney, Flux, Stable Diffusion or a video model.