Visual AI Guide • 12 min read
The Master Midjourney Prompt Guide (v6)
\nPrompting an image generation model is fundamentally different from prompting a text model like ChatGPT. Midjourney does not care about grammar, polite requests, or long conversational explanations. To get breathtaking, professional imagery, you must use a rigid, comma-separated formula based on photographic and artistic terminology.
Want to skip the guide?
Generate your structured prompt instantly using our free tool.
Definition: Prompting an image generation model is fundamentally different from prompting a text model like ChatGPT. Midjourney does not care about grammar, polite requests, or long conversational explanations. To get breathtaking, professional imagery, you must use a rigid, comma-separated formula based on pho
The Visual Prompting Formula
[Subject] + [Environment] + [Lighting] + [Medium/Camera] + [Parameters]
- Subject: The specific focal point (e.g., "A ginger cat wearing aviator goggles").
- Environment: The background and setting (e.g., "sitting in a vintage biplane cockpit").
- Lighting: The mood and illumination (e.g., "golden hour sunlight, cinematic rim lighting").
- Medium/Camera: The artistic style or lens specs (e.g., "shot on 35mm film, f/1.8, highly detailed photography").
- Parameters: Midjourney flags (e.g., "--ar 16:9 --v 6.0").
Midjourney Prompts by Style
Copy these foundational prompts and replace the bracketed text with your specific vision.
Photorealism & Photography
Cinematic Portrait
Cinematic portrait of a [Age/Gender] [Occupation], walking through a [Setting, e.g., rainy Tokyo street], dramatic cyberpunk neon lighting, shallow depth of field, shot on 35mm lens, f/1.8, Kodak Portra 400 film stock, highly detailed, 8k --ar 4:5 --style raw
Product Photography
Commercial product photography of a [Product, e.g., sleek matte black coffee maker], sitting on a [Surface, e.g., absolute white marble countertop], morning sunlight streaming through a window, soft shadows, sharp focus, macro details, minimalist background --ar 16:9
Graphic Design & Illustration
Flat Vector Iconography
Flat vector illustration of a [Subject, e.g., rocket ship launching], minimalist corporate Memphis style, solid pastel background, ui/ux asset, clean lines, no shading, Dribbble aesthetic --ar 1:1 --no gradients, text
Concept Art Environment
Epic concept art of a [Environment, e.g., massive floating ancient city overgrown with glowing vines], fantasy RPG setting, volumetric lighting, atmospheric fog, Greg Rutkowski style, highly detailed digital painting, trending on ArtStation --ar 21:9
Architecture & Renderings
Modern Interior Design
Interior design of a [Room, e.g., sunken conversation pit living room], mid-century modern aesthetic, floor-to-ceiling windows overlooking a forest, travertine floors, walnut wood accents, diffused natural lighting, Architectural Digest photography --ar 16:9
Exterior Architectural Render
Exterior architectural render of a [Building Type, e.g., brutalist concrete gallery], situated in a [Location, e.g., arid desert mesa], golden hour lighting, long dramatic shadows, hyper-realistic Unreal Engine 5 render, octane render, 8k resolution --ar 3:2
Essential Parameters (Flags)
| Parameter | What it does | Example Use Case |
|---|---|---|
| --ar [W:H] | Sets the aspect ratio | --ar 16:9 (YouTube) or --ar 9:16 (TikTok) |
| --no [items] | Negative prompting (excludes things) | --no text, watermarks, ugly |
| --sref [url] | Applies the style of the linked image | --sref https://link.to/image.jpg |
| --stylize [0-1000] | How strongly Midjourney applies its own artistic flair | --stylize 50 (realistic) vs --stylize 800 (artistic) |
| --style raw | Removes Midjourney default beautification | Best for ultra-realistic, unpolished photography |
📌 Key Takeaways
- Prompting an image generation model is fundamentally different from prompting a text model like ChatGPT.
- Midjourney does not care about grammar, polite requests, or long conversational explanations.
- To get breathtaking, professional imagery, you must use a rigid, comma-separated formula based on photographic and artistic terminology.
- The STCO framework (System, Task, Context, Output) provides the most effective structural approach.
- Use AI Prompt Architect to generate structured prompts instantly.
- ⚡Go Pro: Unlimited prompt generations, AI-powered Refine & Analyse, and priority support — from £9.99/mo
Frequently Asked Questions
What is the formula for a good Midjourney prompt?
The ideal Midjourney v6 structure is: [Subject] + [Environment/Setting] + [Lighting] + [Style/Medium] + [Camera/Render specs] + [--parameters]. Unlike language models, Midjourney prefers descriptive noun-phrases over conversational sentences. Avoid words like "create an image of..."
How do I use Midjourney parameters?
Parameters go at the very end of your prompt preceded by two dashes. Essential parameters include --ar (aspect ratio, e.g., --ar 16:9), --v 6.0 (to ensure you are using the latest model), --stylize or --s (how strictly to follow the prompt vs be artistic, scale 0-1000), and --no (negative prompting, e.g., --no text).
Why does Midjourney struggle with text?
While version 6 is much better at generating text than previous versions, diffusion models still struggle because they "draw" text rather than type it. If you need specific text, place it in "quotes" within your prompt. For professional graphics, the best workflow is generating the art in Midjourney and adding the typography in Canva or Photoshop.
What is Midjourney --sref?
Style Reference (--sref) is one of Midjourney's most powerful features. By appending "--sref [URL to an image]" at the end of your prompt, you force the AI to match the visual aesthetic, color grading, and artistic style of that specific image across your new generations.
Generate Prompt Ideas Instantly
Don't know the right camera terminology? Use AI Prompt Architect to generate perfect Midjourney prompt formulas based on simple descriptions.
Build Visual Prompts Free →Midjourney Prompting: The Evidence
Every claim below is sourced from peer-reviewed research and industry reports.Browse all 141 citations →
Few-shot extraction minimizes context window usage vs zero-shot verbose.
3 well-crafted few-shot examples (150 tokens) outperform a 600-token verbose instruction block, saving 75% on input costs per request.
Without concise few-shot examples, developers write lengthy prose instructions that consume 4x more tokens for equivalent or inferior output quality.
Brown et al., 'Language Models are Few-Shot Learners', NeurIPS 2020JSON Schema enforcement eliminates parse errors.
OpenAI structured outputs with JSON Schema achieve 99.9% schema adherence vs <70% with unconstrained generation — a 30x reduction in parse failures.
Without schema enforcement, every 1M requests generate 300K+ malformed responses requiring retries, error handling, and downstream data corruption.
OpenAI, 'Structured Outputs: JSON Schema' documentation, 2024Template systems compress prompt authoring time.
Structured prompt templates cut development time from 4 hours to 20 minutes per prompt (8x reduction) by separating instructions from variables.
Without templates, every new prompt starts from scratch — copying, pasting, and re-debugging the same boilerplate across dozens of prompts.
LangChain, 'Prompt Templates' documentation, 2024Streaming structured data enables progressive rendering.
Streaming JSON objects with Zod validation reduces perceived latency from 3 seconds to 400ms (87% improvement) for AI-powered UI components.
Without streaming, users stare at blank spinners until the full response arrives, creating a sluggish experience that feels broken.
Vercel, 'AI SDK: Streaming Structured Data' documentation, 2024