Midjourney's secret weapon is its parameter system — a small library of flags appended to the end of any prompt that fine-tune everything from aspect ratio to artistic variance. This tutorial is a complete reference and a recipe book. Bookmark it.
Parameters in Midjourney start with two dashes and live at the end of your prompt. Used well, they give you reproducible, precise control. Used carelessly, they will quietly sabotage otherwise excellent prompts (a wrong --style or an unintended --stylize 1000 can flip the aesthetic completely). We will go through each parameter, explain what it does, and show realistic combinations you can copy.
Note: Midjourney evolves quickly. Always check the official Midjourney documentation if a parameter behaves unexpectedly — but the parameters below have been stable across recent versions and are unlikely to change soon.
Think of your prompt as the script and parameters as the production settings. The script tells the model what to make. The parameters tell it how to make it — at what aspect ratio, in which model version, with how much creative variance, and how strongly stylised.
Sets the width:height of the output. Defaults to 1:1. Common useful values:
--ar 1:1 — square, perfect for Instagram posts, app icons, logo concepts.--ar 3:2 — classic photography ratio, good for general use.--ar 4:5 — Instagram portrait, fashion, lifestyle product shots.--ar 16:9 — widescreen landscape, hero banners, YouTube thumbnails.--ar 21:9 — ultra-wide cinematic, great for dramatic scenes.--ar 9:16 — vertical, for Reels, Shorts, TikTok and stories.Selects the Midjourney model version. --v 6 is currently the dominant default for photographic and detailed work. --v 6.1 further sharpens detail. --niji 6 swaps in the Niji model, which is purpose-built for anime and stylised illustration. Always lock a version explicitly in production prompts so updates do not silently shift your aesthetic.
Modifies the version's default behaviour.
--style raw — reduces Midjourney's automatic aesthetic boost. Best for photorealism, product photography, and any time you want literal interpretation.--style expressive / --style cute / --style scenic — only available on Niji, each anchors a different illustration mood.How strongly Midjourney applies its house aesthetic. Range: 0–1000, default 100.
--s 0 — almost no Midjourney "look"; closely follows the prompt literally.--s 100 — default; balanced artistic interpretation.--s 500–750 — high stylisation; richer, more painterly, more creative liberties.--s 1000 — maximum Midjourney aesthetic; gorgeous but sometimes loose with prompt details.How varied the 4-image grid is. Range: 0–100, default 0.
--c 0 — four tightly similar variations.--c 25–50 — meaningful variety, useful for exploration phase.--c 75–100 — wildly different interpretations of the same prompt. Great for brainstorming, dangerous in final production.Pushes the model to make unusual aesthetic choices. Range: 0–3000, default 0. Use sparingly — --w 250 is already noticeable. Excellent for concept art, terrible for product photography.
How much render time is spent. Range: 0.25, 0.5, 1, 2 (defaults to 1). Lower values cost less but produce coarser images; higher values cost more and add detail. --q 0.5 is great for fast exploration; --q 2 is overkill for most cases.
Covered in Topics 11 and 12. Quick reminder: --no excludes elements; --sref imports a style reference; --cref imports a character reference.
Naked prompt — no production controls
cinematic portrait of a samurai in the rain
This will produce a default square (1:1) image at default stylisation, default chaos, on whatever Midjourney version is currently default. If you wanted a 21:9 cinematic crop with raw photographic feel and four similar variations, you will not get it. Reproducibility is also lost — running the same prompt next month may land in a different default.
Production-grade prompt with parameters
Cinematic portrait of a weathered samurai in a torrential
downpour, slow-motion raindrops mid-air, a katana resting
on his shoulder, dramatic chiaroscuro side-lighting from
a lantern off-frame.
Style: photorealistic, shot on Arri Alexa cinema camera,
anamorphic 50mm lens, shallow depth of field, A24 film grade.
--ar 21:9 --v 6 --style raw --s 150 --c 0 --no extra fingers,
deformed hands, watermark, text
The output is now an ultra-wide 21:9 cinematic crop, locked to the v6 model, with raw photographic feel, slightly elevated stylisation, no chaos (so the four-image grid is tightly consistent), and common defects excluded. Reproducible and on-brief.
--v 6 (or whatever you are using). Production prompts should never rely on the default.--style raw for photography and product work. Skip it for editorial illustration or fantasy.--s 100 is balanced. Lower for literal prompts, higher for artistic interpretations.--c 0 for production runs, --c 25–50 when you are still exploring directions.--no for exclusions, --sref or --cref for image references.Tip: Build a "parameter footer" template you paste at the end of almost every prompt — something like
--ar 16:9 --v 6 --style raw --s 150 --no text, watermark, deformed hands. Override only what the specific brief needs.
Generate the same prompt with --s 50, --s 250, and --s 750. Side-by-side, observe how stylisation shifts from literal to painterly. Find your personal default.
Take any prompt and run it at --c 0, --c 25, and --c 75. See how the four-image grid becomes more diverse. Decide which chaos level matches your current task.
Build your personal "parameter footer" and save it in your notes. Use it across five different prompts this week and note where you override. Within a week you will have a tuned default that fits your workflow.
--v in production — relying on the default lets future model updates silently change your aesthetic.--ar picks the canvas; --style raw tones down Midjourney's artistic boost; --s controls stylisation strength.--c (chaos) controls diversity inside the 4-image grid — high for brainstorming, zero for production.Sign in to join the discussion and post comments.
Sign inPrompt Engineering for Content & Copywriting
Write blogs, ads, emails, and social media content ten times faster with AI. 13 practical tutorials on prompt engineering for content creators and copywriters.
Prompt Engineering for Education & Learning
Use AI as your personal tutor. Learn how to study faster, create lesson plans, generate practice questions, master languages, and prepare for competitive exams with smart prompts.
Prompt Engineering for Specific AI Tools
Tool-by-tool mastery — deep dives into ChatGPT, Claude, Gemini, GitHub Copilot, Midjourney, Stable Diffusion, and more. Learn the exact prompting techniques each platform rewards.
Prompt Engineering for Developers
Use AI as your coding co-pilot. 18 tutorials on writing prompts to generate clean code, debug faster, write tests, build APIs, and ship better software.
Prompt Engineering for Business & Productivity
Use AI to work smarter — automate tasks, make better decisions, and communicate professionally. 12 practical business prompt tutorials for professionals.
Prompt Engineering Projects & Real-World Applications
Twelve hands-on projects that turn prompt engineering theory into a portfolio. Build chatbots, content generators, RAG systems, and more.