AI image prompt generator

Run on September 28, 2026: 5 inputs on 15 models, each with this tool's own instructions.

Turn this idea into a detailed prompt for an AI image generator: subject, setting, style, lighting, camera and mood, then an "Avoid:" line. Make the style a digital illustration.

---
a poster for a jazz night called Blue Hour, Friday 14 November, 8 pm, at The Lantern

A promotional poster for a jazz night titled “Blue Hour,” with the event details “Friday 14 November, 8 pm” and “at The Lantern” clearly readable and spelled exactly. Set the scene in an intimate jazz club with a saxophone and softly suggested stage lights, rendered as a polished digital illustration. Use moody blue and indigo tones with warm amber accents, dramatic low lighting, and a balanced portrait composition with bold, legible typography. Evoke a cool, soulful, late-night mood.

Avoid: blurry details, misspelled or extra text, clutter, distorted instruments, illegible typography.

A vertical jazz-night poster with elegant lettering reading Blue Hour, Friday 14 November, 8 pm, at The Lantern, centred on a silhouetted saxophonist beneath an art-deco streetlamp, swirling smoke, and tiny musical notes, set against a moody midnight-blue cityscape, rendered as a polished digital illustration with subtle grain. Soft cyan and amber club lighting, deep indigo shadows, warm gold accents; symmetrical composition with bold negative space, slightly low angle. Elegant, intimate, nocturnal mood.
Avoid: blur, extra fingers, clutter, warped proportions, misspelled or garbled text, watermark, harsh daylight, low contrast.

1. Write your idea in a few words, or a sentence: what, where, and the feeling. Up to 800 characters.

2. Pick a style and the shape of the picture.

3. Press “Write the prompt”. Without an account you see its start (1 free run a day, no account needed).

4. Copy the prompt into your image generator. Signed in on a paid plan, make the image in llmwise.

• Say where it'll be used

  Say what the picture is for (a blog header, a product listing, a phone wallpaper) and the prompt picks a composition that fits.

• Change one thing at a time

  Generate once, then change one thing at a time: the light, the lens, the time of day. You'll learn what each part of the prompt does.

• Name the mood

  A reference to a mood (“quiet, early morning”) steers colour and light better than a list of colours.

• Don't mention what you don't want

  Image models draw what you mention. Leave out what you don't want in the picture, and put it on the Avoid line instead.

• The subject, first and specific

  The subject comes first and is specific: “a steel water bottle with a bamboo cap” rather than “a bottle”. Image models weigh the start of a prompt most.

• Light and camera, named

  Light and camera make a picture look intentional: “soft window light from the left, 50 mm lens, shallow depth of field” does more than “beautiful”.

• Nothing of your idea lost

  Everything in your idea should still be there. A generator that swaps your rooftop for a beach, or your family for one person, isn't describing your picture.

• Short text, in quotes

  Most image models can't spell reliably. If the picture needs words on it, keep them short and put them in quotes in the prompt.

• Without an account: 1 free run a day, on GPT-6 Luna, shared with the free message on our home page. It takes up to 800 characters, writes up to about 600 tokens (roughly 450 words), and shows you the start of the image prompt.

• Signing up is free, with an email link and no card. The image prompt you ran comes with you as your first chat, and you get 5 free messages on almost every model (all but Claude Fable 5.1, Claude Opus 5.5, and GPT-6 Astra).

• Signed in, the tool runs on the model you pick, on your plan, shows the whole answer, and you can carry on in the chat with any model, or ask another model the same thing.

More free tools are on the tools page, and what each plan gives is on pricing.