ChatGPT Midjourney Portrait Prompt Generator Prompt
You are a professional AI art director who creates Midjourney prompts resulting in editorial-quality portraits.
Category
🎨 Image AI
Difficulty
Intermediate
Models
3
Last Updated
2026-06-28
Works with
📄 Example output
⚠️ Common Mistakes
❓ FAQ
⚙️ Fill in your variables
📋 Prompt
You are a professional AI art director who creates Midjourney prompts resulting in editorial-quality portraits.
Subject: [subject description]
Style: [style — photorealistic/painterly/cinematic/fashion editorial/environmental portrait]
Lighting: [lighting — golden hour/studio/dramatic chiaroscuro/soft diffused/neon]
Mood: [mood — confident/mysterious/joyful/contemplative/powerful]
Task: Create 4 Midjourney portrait prompt variations:
FORMULA:
[subject], [environment/context], [style], [lighting], [camera/lens], [mood/atmosphere], [quality tags] --ar [ratio] --v 6.1 --stylize [value]
Variation 1 — Photorealistic:
/imagine prompt: [full prompt] --ar 2:3 --v 6.1 --stylize 100
Variation 2 — Cinematic:
/imagine prompt: [full prompt] --ar 16:9 --v 6.1 --stylize 250
Variation 3 — Editorial/Fashion:
/imagine prompt: [full prompt] --ar 4:5 --v 6.1 --stylize 400
Variation 4 — Artistic:
/imagine prompt: [full prompt] --ar 1:1 --v 6.1 --stylize 750
For each: explain what the --stylize value does to the result.
Include a negative prompt (--no) for all variations.
Subject: [subject description]
Style: [style — photorealistic/painterly/cinematic/fashion editorial/environmental portrait]
Lighting: [lighting — golden hour/studio/dramatic chiaroscuro/soft diffused/neon]
Mood: [mood — confident/mysterious/joyful/contemplative/powerful]
Task: Create 4 Midjourney portrait prompt variations:
FORMULA:
[subject], [environment/context], [style], [lighting], [camera/lens], [mood/atmosphere], [quality tags] --ar [ratio] --v 6.1 --stylize [value]
Variation 1 — Photorealistic:
/imagine prompt: [full prompt] --ar 2:3 --v 6.1 --stylize 100
Variation 2 — Cinematic:
/imagine prompt: [full prompt] --ar 16:9 --v 6.1 --stylize 250
Variation 3 — Editorial/Fashion:
/imagine prompt: [full prompt] --ar 4:5 --v 6.1 --stylize 400
Variation 4 — Artistic:
/imagine prompt: [full prompt] --ar 1:1 --v 6.1 --stylize 750
For each: explain what the --stylize value does to the result.
Include a negative prompt (--no) for all variations.
SUBJECT: 'Young South Asian woman, architect, late 30s'
STYLE: Cinematic environmental portrait
LIGHTING: Late afternoon golden hour
MOOD: Confident and contemplative
VARIATION 1 — PHOTOREALISTIC:
/imagine prompt: portrait of a South Asian woman architect in her late 30s, standing in front of an under-construction building site, wearing smart professional attire, golden hour sunlight casting warm shadows, Canon 85mm f/1.4 lens, shallow depth of field, confident posture looking directly at camera, architectural blueprints visible in background, photorealistic photography, editorial portrait style, skin texture detail, natural highlights --ar 2:3 --v 6.1 --stylize 100 --no cartoon, illustration, painted, blurry, distorted features, extra fingers
[Stylize 100 explanation: Follows your prompt closely. Realistic result. Very little 'artistic interpretation'.]
VARIATION 2 — CINEMATIC:
/imagine prompt: cinematic portrait of South Asian female architect, 30s, dramatic golden hour backlight creating rim lighting silhouette, modern construction site background with steel beams, professional urban clothing, film still aesthetic, anamorphic lens flare, rich colour grading, atmospheric haze, confident expression, architectural drawing partially visible, directed by Roger Deakins cinematography style --ar 16:9 --v 6.1 --stylize 250 --no cartoon, anime, low quality, watermark
[Stylize 250: Balances your prompt with Midjourney's aesthetic sense. Slightly more dramatic than Stylize 100.]
STYLE: Cinematic environmental portrait
LIGHTING: Late afternoon golden hour
MOOD: Confident and contemplative
VARIATION 1 — PHOTOREALISTIC:
/imagine prompt: portrait of a South Asian woman architect in her late 30s, standing in front of an under-construction building site, wearing smart professional attire, golden hour sunlight casting warm shadows, Canon 85mm f/1.4 lens, shallow depth of field, confident posture looking directly at camera, architectural blueprints visible in background, photorealistic photography, editorial portrait style, skin texture detail, natural highlights --ar 2:3 --v 6.1 --stylize 100 --no cartoon, illustration, painted, blurry, distorted features, extra fingers
[Stylize 100 explanation: Follows your prompt closely. Realistic result. Very little 'artistic interpretation'.]
VARIATION 2 — CINEMATIC:
/imagine prompt: cinematic portrait of South Asian female architect, 30s, dramatic golden hour backlight creating rim lighting silhouette, modern construction site background with steel beams, professional urban clothing, film still aesthetic, anamorphic lens flare, rich colour grading, atmospheric haze, confident expression, architectural drawing partially visible, directed by Roger Deakins cinematography style --ar 16:9 --v 6.1 --stylize 250 --no cartoon, anime, low quality, watermark
[Stylize 250: Balances your prompt with Midjourney's aesthetic sense. Slightly more dramatic than Stylize 100.]
🏆
💡 Pro Tips
Best model for this prompt
ChatGPT
ChatGPT (GPT-4o / GPT-5)
Add the camera and lens (e.g., 'shot on Canon 5D Mark IV, 85mm f/1.4') — this dramatically improves portrait realism in Midjourney v6
Use 'natural skin texture' and 'subsurface scattering' for more realistic skin in photorealistic styles
Stylize values: 50–150 = follows prompt closely, 250–500 = artistic interpretation, 600–1000 = maximum Midjourney style expression
Reference a specific photographer or cinematographer ('in the style of Annie Leibovitz' or 'Gregory Crewdson lighting') for distinctive results
Forgetting --no for unwanted elements — Midjourney often adds watermarks, extra people, or distorted hands without this
Too vague a subject description — 'beautiful woman' generates stock-photo aesthetics; specific descriptions create unique results
Stacking too many styles — 'photorealistic painting cinematic anime portrait' confuses the model; pick one dominant aesthetic
Ignoring aspect ratio — --ar 2:3 (portrait) and --ar 16:9 (landscape) produce very different compositions even with identical prompts
- What version of Midjourney should I use?Midjourney v6.1 (released 2025) is the current best for portraits and photorealism. Niji 6 is the best option for anime and illustration styles. Always specify --v 6.1 or --niji 6 in your prompts.
- How do I fix hands and fingers in Midjourney?Add --no deformed hands, extra fingers to your negative prompt. Alternatively, crop hands out of the composition by specifying 'headshot' or 'waist-up portrait'. The Vary (Region) feature in Midjourney can also fix specific problem areas.
- Can this prompt work for DALL-E 3?Use the Variation 1 prompt as natural language for DALL-E 3 — remove the Midjourney parameters (--ar, --v, --stylize, --no) and convert to complete sentences. DALL-E 3 understands natural language better than parameter-based instructions.
- What's the --stylize parameter and how should I use it?--stylize (or --s) controls how much Midjourney applies its own aesthetic training vs. following your prompt literally. Range: 0–1000. For portraits, 100–250 is the sweet spot for professional-looking results. Below 50 looks unpolished; above 600 is highly stylised.