Midjourney Prompts Beginner 2 min read

DALL·E 3 Prompt Examples: Scene, Light and Text Control

DALL·E 3 handles natural-language sentences well; these templates pin down scene, light and text in one paragraph.

Works with: DALL·E Updated: 24 October 2026
Prompt 1091 characters · 3 variables
You are writing prompts for DALL·E 3, which reads full sentences better than keyword lists.

IMAGE
[WHAT THE IMAGE SHOWS]
USE: [USE CASE]
TEXT IN IMAGE: [EXACT TEXT OR "none"]
STYLE: [PHOTOGRAPH / ILLUSTRATION / 3D / FLAT]

WRITE ONE PARAGRAPH THAT COVERS
- The scene in plain sentences: what is where, who is doing what.
- The light: where it comes from, what it does to the subject.
- The style: medium, level of detail, colour palette.
- The framing: wide, close, top-down.
- Any text, quoted exactly, with its position and font character ("bold sans-serif, centred").

RULES
- Use complete sentences, not comma lists.
- Do not ask for more than one scene per prompt.
- If text is required, keep it under 5 words and quote it exactly; longer text breaks.
- Avoid artist names; describe the style instead.

OUTPUT
A) MAIN PROMPT — one paragraph, 80-120 words.
B) 3 VARIATIONS — same scene, different light / different framing / different style.
C) TEXT-ONLY VERSION — if text matters, a prompt optimised for legibility.
D) WHAT WILL BREAK — the two elements DALL·E usually gets wrong here.

Describe the picture the way you would describe it to a person, then add the technical bits.

02 — Variables

Fill in the brackets

The highlighted fields are the places where your own details go. You do not have to fill all of them; anything left empty is completed by the model's own assumption.

[WHAT THE IMAGE SHOWS]
[USE CASE]
[PHOTOGRAPH / ILLUSTRATION / 3D / FLAT]

03 — How to use

How to use it, step by step

  1. 01

    DALL·E follows sentences, so write prose. Keyword stacks work against it.

  2. 02

    Keep on-image text under five words and quote it exactly; longer strings turn into nonsense.

  3. 03

    Iterate by editing one sentence, not by rewriting the whole prompt.

04 — Variations

Same prompt, different result

One prompt does not solve every case. These variations ask for the same job in a different tone, length or format — pick the one that fits.

Infographic

Same structure for a simple infographic: 3 labelled elements, flat colours, generous whitespace, minimal text.

Book cover

Same structure for a cover: title area left clear, strong silhouette, single focal point.

05 — Tips

Notes that take the result up a level

  • If the image must contain exact text, generate the visual without text and add typography in a design tool.
  • DALL·E tends to centre and fill the frame. Say "generous negative space on the left" if you need room for text.

06 — FAQ

Frequently asked about Midjourney Prompts

Does DALL·E do text well?

Better than most, but still unreliable past a few words. Short quoted words work; sentences do not.

Why do prompts in keyword style fail?

DALL·E is trained on natural captions. Full sentences give it relationships between objects; keyword lists give it a bag of nouns.

Next

You may also find these useful

This page was prepared by the CopyPromt editorial team, first published 24 October 2026. Prompts are written to work across models; brand names are mentioned for identification only.