Productivité & Carrière

Midjourney Pro Visuals in Practice: The Code and Commands That Really Matter

Midjourney Pro Visuals: the essentials in one article — real code, diagrams and concrete steps, excerpts from a 43-lesson course.

REHOUMA Haythem

• 10 min read
Midjourney Pro Visuals in Practice: The Code and Commands That Really Matter

No endless theory here: open the terminal and practice. Here is the essence of Midjourney Pro Visuals, extracted directly from a complete 43-lesson course — with real code you can copy-paste right now.

tl;dr
  • Introduction and Setup
  • Anatomy of a Good Prompt
  • Advanced Parameters and Flags
  • Styles and Aesthetics
  • Character Consistency and Style Reference
~$ cat ./parcours.md # Midjourney Visuels Pro — 10 chapters
01
Introduction and Configuration
→ Course presentation and Midjourney→ Subscriptions, Discord vs Web App+ 1 more lessons
02
Anatomy of a Good Prompt
→ The 6 components of an effective prompt→ Visual description vocabulary+ 2 more lessons
03
Advanced Parameters and Flags
→ Ratio (--ar) and resolutions→ Stylization (--s) and raw mode+ 2 more lessons
04
Styles and Aesthetics
→ Pro photography→ Illustration and vector+ 2 more lessons
05
Character Consistency and Style Reference
→ Character Reference (--cref)→ Style Reference (--sref)+ 1 more lessons
06
Vary Inpainting and Retouching
→ Vary Strong, Subtle and Region→ Pan and Zoom Out+ 2 more lessons
07
Upscale and Preparation for Printing
→ Upscale Subtle vs Creative→ External upscalers+ 1 more lessons
08
Pro Usage Marketing Branding and Content
→ Social media visuals→ Product mockups and e-commerce+ 2 more lessons
🏁
Final project (+ 2 chapters along the way)
→ You leave with a concrete and demonstrable project

Style Reference (--sref)

NOTEObjective — Apply the same visual atmosphere (colors, texture, rendering) to several different images using --sref (style reference), and adjust its influence with --sw.

Learning objectives

TIPBy the end of this module
  • Distinguish style reference from character reference
  • Use --sref with one or more URLs
  • Adjust influence with --sw
  • Use numeric style codes
  • Create brand visual consistency

The intuition: copy the atmosphere, not the subject

While --cref copies a character, --sref copies a style: the color palette, texture, lighting, and overall “grain” of a reference image — but not its subject. You can therefore generate a car, a landscape, and a portrait that all share the same visual atmosphere.

It is the key tool for brand consistency: all your visuals “look alike” without being identical.

The --sref syntax

Numeric code

A reusable style identifier. Useful for replaying an atmosphere discovered at random.

Brand consistency

The typical pro scenario: you define ONE “house” style reference and apply it to all your visuals.

WARNINGCaution: do not combine too many style references at once (3+). The model averages the atmospheres and the result becomes blurry and inconsistent. One or two well-chosen references are enough.

Pro photography

NOTEObjective — Produce convincing photo-realistic images in the three main genres: portrait, landscape, and product photography, using the technical vocabulary of the trade.

Learning objectives

TIPBy the end of this module
  • Write a credible portrait prompt
  • Compose a landscape with depth
  • Create a studio-level product photo
  • Use the right lens and lighting terms for each genre
  • Activate raw mode to maximize realism

The intuition: imitate real-world craft

To obtain a “real photo,” you must speak like a real photographer: specify the lens, lighting, and type of shot. Midjourney has seen millions of real photos captioned with these terms. The more you describe a credible photographic setup, the more authentic the result appears.

The portrait

A good portrait combines a long focal-length lens (which flatters features), shallow depth of field (blurred background), and soft light.

First image with /imagine

NOTEObjective — Generate your very first image with the /imagine command, understand the four-image grid, and learn the U, V, and regenerate buttons.

Learning objectives

TIPBy the end of this module
  • Use the /imagine command correctly
  • Read and interpret the four-image grid
  • Distinguish the U (Upscale) and V (Variation) buttons
  • Know the essential beginner commands
  • Complete your first render from start to finish

The intuition: one command, four proposals

Everything starts with /imagine. You type this command followed by your description, and Midjourney thinks for about a minute before returning a grid of four images numbered 1, 2, 3, 4 (left to right, top to bottom).

Why four? Because generation is partly random: the model explores four different interpretations of your prompt. You then choose the one that speaks to you most, then refine or enlarge it.

Your first prompt

On Discord, in a channel where the Midjourney bot is present (or in a direct message to the bot), type:

TIPTip: on the web app it is even simpler: an input field at the top of the screen replaces the /imagine command. Just type the description and press Enter.

Reading the grid: the U and V buttons

Buttons appear beneath the grid. Here is what they do.

ButtonActionWhen to use
U1 to U4Upscale: isolates and enlarges the chosen imageWhen one of the four images suits you
V1 to V4Variation: generates 4 new versions close to the chosen imageWhen an image is almost good but needs adjustment
Re-roll (circular arrow)Completely restarts the same promptWhen none of the four images work

U = Upscale

The U button selects an image and processes it at full resolution. This is the step before downloading.

V = Variation

The V button explores close alternatives: same spirit, different details. Ideal for converging on the right result.

Essential commands to know

Beyond /imagine, a few commands come up constantly. Here is your basic toolkit.

WARNINGCaution: do not download the four-image grid directly (it is low resolution). Always go through an Upscale (U button) before saving, otherwise your image will be small and blurry.
go-further

This article covers the most useful excerpts — the complete Midjourney Pro Visuals course (11 chapters, 43 lessons, corrected exercises and final project) takes you all the way.

./access-the-full-course free course: Claude Cowork

FAQ

How long does it take to learn Midjourney Pro Visuals?
With structured progression (11 chapters, 43 short and practical lessons), you reach an operational level in a few weeks at 30–60 minutes per day. The key is to practice each concept immediately.
Are any prerequisites required?
Basic computer skills are enough. If you can use a terminal and read simple code, you are ready.
Where should I start concretely?
Reproduce the commands in this article, then follow the complete Midjourney Pro Visuals course: it sequences the 43 lessons in order, with exercises and a final project.
./read-also
→ Get started with Excel Copilot: your first concrete step today→ Essential AI Toolbox: the 9 key steps to go from zero to operational→ AI Career Opportunities explained simply (with diagrams and real code)

📬 Want to receive this type of guide every week? Subscribe for free — real code, zero fluff.

Stay up to date

Enter your email
Subscribe