AI Influencer Prompts: 50+ Templates & Examples for Every Tool

Tool-by-tool at a glance

The difference between an image that screams "AI-generated" and one that stops scrollers mid-swipe? Usually 15 well-chosen words.

Whether you're creating your first AI character or refining one that already has followers, the templates below will save you hours of trial and error. Use the interactive generator to build custom prompts for your tool, then explore the full guide for 50+ copy-paste templates and a negative prompts cheat sheet.

The Anatomy of a Perfect AI Influencer Prompt

Every high-performing prompt follows the same structure, regardless of tool. Think of it as giving a photographer a creative brief — the more specific your direction, the better the shot.

[Subject] + [Physical Details] + [Clothing/Style] + [Action/Pose] + [Setting/Location] + [Lighting] + [Camera/Technical] + [Mood/Atmosphere]

1. Subject

Start with what you're generating. Be specific about age, gender, and framing.

• ❌ Weak: "a woman"
• ✅ Strong: "a 26-year-old woman, medium shot, looking at camera"

2. Physical Details

Define the features that make your character recognizable. Consistency starts here.

Face: eye color, eye shape, skin tone, freckles, dimples, jawline. Hair: color, length, texture, style. Build: athletic, slim, curvy, petite. Write a "character sheet" prompt once and reuse it — see our Face Consistency Guide for methods.

3. Clothing & Style

Clothing tells the story of your influencer's niche. Fashion: "oversized cream blazer, gold jewelry." Fitness: "matching sage activewear set, sports watch." Travel: "linen shirt, straw hat, leather sandals."

4. Action & Pose

Add life: "laughing while looking over her shoulder," "mid-stride on a crosswalk, wind catching her hair," "leaning against a railing, holding a coffee cup."

5. Setting & Location

"Sunlit café terrace in Barcelona," "modern minimalist apartment with floor-to-ceiling windows," "busy Tokyo street at night with neon reflections."

6. Lighting

Golden hour, soft natural light, studio lighting, neon/ambient, rim light. Lighting is what separates amateur-looking images from professional ones.

7. Camera & Technical

Canon EOS R5, 85mm f/1.4 (portrait king), 35mm (lifestyle), shallow depth of field. Film stocks: Kodak Portra 400, Fuji Superia.

8. Mood & Atmosphere

"Candid and authentic feel," "editorial and aspirational," "cozy and intimate," "bold and confident energy."

Midjourney (V7)

Midjourney V7 brought dramatically improved realism with its RealFace engine — better skin texture, natural pores, and finally, proper hands. Prompt style: Descriptive phrases, comma-separated. Less is more — V7 understands context better than V6.

Formula:

[subject description], [clothing], [action/pose], [setting], [lighting], [camera reference], [mood] --ar 4:5 --style raw --s 80 --v 7

Example:

26-year-old woman with long wavy brunette hair and hazel eyes, wearing a cream knit sweater and gold necklace, laughing candidly in a cozy café, warm window light on her face, shallow depth of field, Canon EOS R5, 85mm f/1.4, authentic lifestyle photography --ar 4:5 --style raw --s 80 --v 7

Pricing: $10–60/month. Read our Midjourney review.

Flux / Flux 2

Flux understands natural language better than any competitor — write prompts like you'd describe a photo to a friend. Prompt style: Full sentences. No keyword spam. Quality tags like "masterpiece" are unnecessary.

Formula:

A [shot type] of a [age]-year-old [gender] with [physical details]. [He/She] is [action] in [setting]. [He/She] is wearing [clothing details]. The lighting is [lighting description], creating [mood]. Shot on [camera], [lens], [aperture].

Example:

A medium shot of a 26-year-old woman with long wavy brunette hair and warm hazel eyes. She is sitting in a sunlit café by the window, laughing naturally while holding a ceramic coffee cup. She is wearing a cream knit sweater and delicate gold necklace. The soft morning light creates gentle shadows on her face, with visible skin texture and natural imperfections. Shot on Canon EOS R5, 85mm lens at f/1.4, shallow depth of field with warm bokeh in the background.

Pricing: Free (Dev), ~$0.04/image (Pro API).

Stable Diffusion / SDXL

More control, more complexity. SDXL responds better to natural language than SD 1.5, but tags still help. Prompt style: Tag-based with weighted keywords. Use positive AND negative prompts.

Weighting: (important word:1.3), (less important:0.7). Use LoRA for face consistency — Face Consistency Guide.

Example prompt (SDXL):

photorealistic portrait of a 26-year-old woman, long wavy brunette hair, hazel eyes, (natural skin texture, visible pores:1.2), wearing cream knit sweater and gold necklace, sitting in sunlit café, warm window light, shallow depth of field, (Canon EOS R5, 85mm f/1.4:1.1), candid authentic expression, lifestyle photography

Negative prompt (essential):

(worst quality, low quality:1.4), bad anatomy, bad hands, deformed, extra fingers, missing fingers, poorly drawn face, mutation, blurry, watermark, text, logo, signature, overexposed, underexposed, painting, cartoon, 3d render, anime, illustration, CGI, disfigured, ugly, tiling, cross-eyed

Read our Stable Diffusion review.

Leonardo AI

The most beginner-friendly platform. PhotoReal V2 and built-in presets do most of the work. Prompt style: Descriptive, natural language. Use "Improve Your Prompt" to auto-enhance.

Example:

Medium shot of a 26-year-old woman with long wavy brunette hair and hazel eyes, wearing a cream knit sweater and gold necklace, sitting in a sunlit café, warm natural window light with gentle shadows, Canon EOS R5, 85mm portrait lens, shallow depth of field, candid authentic expression, photorealistic skin texture with visible pores, lifestyle photography

Negative prompt:

blurry, distorted, extra fingers, bad anatomy, watermark, text, cartoonish, painting, 3d render, overexposed, plastic skin

Read our Leonardo AI review.

Affogato (formerly RenderNet)

Purpose-built for AI influencers. FaceLock handles face consistency automatically — your prompts focus on scenes and styling only. Upload one face photo, generate unlimited images with that face.

Example (with FaceLock enabled):

wearing a cream knit sweater and gold necklace, sitting in a sunlit café, warm window light, laughing candidly, holding coffee cup, shallow depth of field, lifestyle photography, photorealistic, 8K, ultra-detailed

Negative prompt:

nsfw, deformed, extra limbs, bad anatomy, deformed pupils, text, worst quality, jpeg artifacts, ugly, duplicate, morbid, mutilated

Read our RenderNet/Affogato review.

Negative Prompts Cheat Sheet

Essential for Stable Diffusion, recommended for Leonardo, available in Midjourney via --no. Flux handles most issues natively.

Universal (copy-paste):

bad anatomy, bad hands, deformed, extra fingers, missing fingers, poorly drawn face, blurry, watermark, text, logo, signature, overexposed, underexposed, low quality, worst quality, jpeg artifacts, duplicate, morbid, mutilated, extra limbs, disfigured, ugly

Photorealism (enhanced):

cartoon, 3d render, anime, illustration, CGI, painting, oil painting, sketch, drawing, digital art, artstation, unreal engine, plastic skin, waxy, overprocessed, oversaturated, artificial, unrealistic, bad photography, bad composition, harsh lighting

Portrait-specific:

cross-eyed, asymmetric eyes, extra eyes, fused face, cloned face, poorly rendered teeth, unnatural mouth, dead eyes, flat eyes, deformed iris, deformed pupils, long neck, merged faces

50+ Copy-Paste Prompt Templates

Replace [brackets] with your character's details. Full list in the guide; here are a few essentials.

Instagram Feed (--ar 4:5)

Casual lifestyle:

[age]-year-old [gender] with [hair description] and [eye color], wearing [casual outfit], [casual action] in [everyday setting], soft natural lighting, Canon EOS R5, 50mm f/1.8, candid authentic feel, lifestyle photography

Fashion OOTD:

Full body shot of [age]-year-old [gender] with [hair], wearing [outfit description], standing [pose] in [urban location], golden hour lighting, street style photography, 85mm lens, shallow depth of field

Stories & Reels (--ar 9:16)

Mirror selfie:

[age]-year-old [gender] with [hair], taking a mirror selfie in [setting], wearing [outfit], holding phone at chest level, natural room lighting, iPhone 15 Pro photo, candid and casual

Brand deal / Sponsored

Product showcase:

[age]-year-old [gender] with [hair], holding [product] at face level with [natural expression], [setting], professional studio-quality lighting, commercial photography, 85mm macro detail on product

For 20+ more templates (fitness, travel, beauty, seasonal, niche-specific), use the generator above or adapt these with the modifiers reference below.

Prompt Modifiers Quick Reference

Common Prompt Mistakes (And How to Fix Them)

• Keyword stuffing: ❌ "8K, HDR, masterpiece, best quality" → ✅ "photorealistic portrait, Canon EOS R5, 85mm f/1.4, natural skin texture"
• Plastic skin: ❌ "flawless skin, perfect face" → ✅ "natural skin texture, visible pores, slight imperfections, catchlight in eyes"
• Aspect ratios: Use 4:5 for feed, 9:16 for Stories/Reels, 16:9 for YouTube thumbnails, 2:3 for Pinterest.
• Prompt order: Subject first, then details. Most models process front-to-back.
• Lighting: ❌ "good lighting" → ✅ "warm golden hour sunlight from the left, soft fill"
• Story: ❌ "woman standing in a room" → ✅ "woman leaning against kitchen counter, mid-laugh, holding half-eaten apple"

Frequently Asked Questions

Do I need different prompts for each AI tool?

Yes. Midjourney uses comma-separated descriptive phrases with parameters (--ar, --style raw). Flux prefers full natural-language sentences. Stable Diffusion works best with weighted tags and requires separate negative prompts. Leonardo AI is forgiving with natural language. Affogato prompts focus on scene description since FaceLock handles the face.

How do I keep my AI influencer's face consistent across images?

Prompts alone won't guarantee perfect face consistency. Use tool-specific features: Midjourney's --cref parameter, Leonardo's Character Reference, Affogato's FaceLock, or train a custom LoRA model. Read our full Face Consistency Guide for detailed methods.

What's the best aspect ratio for AI influencer content?

For Instagram feed: 4:5 (1080x1350px). For Stories, Reels, and TikTok: 9:16. For YouTube thumbnails: 16:9. For Pinterest: 2:3. Always generate at the platform-native ratio — cropping after generation loses quality.

Should I mention "photorealistic" or "8K" in my prompts?

For Flux: no, it defaults to quality output. For Midjourney V7: not necessary, use --style raw instead. For Stable Diffusion: include "photorealistic" but skip generic quality tags. For Leonardo: enable PhotoReal V2 instead of relying on text. Specific camera references (Canon EOS R5, 85mm lens) work better than vague quality keywords across all tools.

How long should my prompts be?

For Midjourney: 40-75 words hits the sweet spot. Flux: 50-100 words works well since it understands complex sentences. Stable Diffusion: 30-60 words of weighted tags. Leonardo AI: 30-60 words, their "Improve Prompt" feature expands short inputs. Longer isn't always better — focus on specific, meaningful details over word count.

Can I use the same prompt across different tools?

You can use the same concept, but you should adapt the format. Take a Midjourney prompt and rewrite it as natural language for Flux, or convert it to weighted tags for Stable Diffusion. The templates in this guide use a universal format you can quickly adapt.