DALL·E prompt generator · free

DALL·E prompt generator

DALL·E takes plain instructions, not parameters. Upload an image and get one clear sentence-style prompt that describes it end to end.

Free & instant — no sign-up to get prompts

A cinematic portrait being decoded into a prompt
Extracted prompt · Midjourneycinematic portrait, soft warm window light, muted earth tones, 85mm, shallow depth of field --ar 2:3 --style raw

The workbench

Upload once. Get every model's version.

One analysis, formatted for the model you actually use — copy it, or generate the image right here.

Your image

Portrait preview

Or try one

General
Midjourney
Flux
Stable Diffusion
DALL·E
Gemini · Nano Banana
ChatGPT
A cinematic, waist-up portrait of a young woman in a beige trench coat, softly lit by window light with muted earth tones and a shallow depth of field.

Subject

woman, beige trench coat

Composition

waist-up, centered

Lighting

soft window light

Style

cinematic, muted

Camera

85mm, f/1.8

Negative

blur, extra fingers

Free · no login

Proof

The prompt rebuilds your image.

Every prompt comes with a recreation, side by side with your original — so you know it works before you use it.

Original uploadGenerated from the prompt
originalgenerated ✦
Original uploadGenerated from the prompt
originalgenerated ✦
Original uploadGenerated from the prompt
originalgenerated ✦

Prompt library

Steal from a gallery of working prompts.

Browse by look, model and subject. Tap any image to load its prompt into the workbench.

From image to DALL·E prompt

What a DALL·E prompt generator does

It turns your reference into one clear instruction — the format DALL·E takes, with no parameter flags and no negative prompt to fill in.

DALL·E is the outlier among image models in how little syntax it wants. There are no trailing parameters, no weighting brackets and no separate negative field; there is one instruction, and the system will often expand it before rendering. That expansion is why vague prompts drift so far from the reference — anything you leave unsaid gets filled in for you. The prompt this page produces is therefore written to be specific and self-contained: the subject stated plainly, then the framing, the light, the palette and the finish, in one connected instruction that leaves little room for reinterpretation. Upload your image, read the fields it extracted, edit anything that is off, and copy a single block you can paste straight into DALL·E.

DALL·E specifics

Writing for a model that rewrites you

The main failure mode with DALL·E is not a bad prompt — it is an underspecified one.

One instruction, no flags
The prompt is a single descriptive instruction. Parameter syntax borrowed from Midjourney or weighting brackets borrowed from Stable Diffusion are not included, because DALL·E treats them as literal words in your request.
Specific enough to survive rewriting
Because the system commonly expands a short prompt before generating, the output names framing, light direction, palette and finish explicitly — the attributes that otherwise get invented for you.
No negative prompt
There is nowhere to put exclusions, and phrasing them as "no X" tends to summon X. Anything worth avoiding is expressed as what should be there instead.
Shape stated in words
Aspect ratio is requested in plain language — a portrait, landscape or square frame — since there is no ratio parameter to set.

Other models

Same picture, other models

The same upload produces a prompt for each model. For ChatGPT, Midjourney and Flux the wording differs — parameters, tag weighting and prompt length all change what each model does with the same description. Compare them all from the home page.

DALL·E questions

DALL·E prompt questions

Why does DALL·E change my prompt?
The system commonly rewrites a request before rendering it, partly to add detail and partly for safety. You cannot switch that off, but you can reduce its effect: the more specific your prompt is about framing, light and finish, the less room there is for the rewrite to invent something else. That is exactly what the prompt on this page is built to do.
Can I add --ar or other parameters?
No. DALL·E has no parameter syntax and will read those characters as part of your description. Ask for the shape in words instead — the generated prompt already does.
Why was my prompt refused?
Requests naming living artists, recognisable public figures or trademarked characters are commonly declined. If a refusal happens, remove the name and keep the visual description — the style attributes usually get you most of the way without it.
Is the prompt free?
Yes. Reading and copying it needs no account. An account is only required to generate a recreation image on this page.

What's the prompt behind that?

Drop an image in and find out. Free, no sign-up.

Read a prompt