THE CRAFT · TOOL SELECTION
Model Guides
Around twenty models and studios with their optimal prompt format — ChatGPT Image 2.0, Seedance, Kling, FLUX, Midjourney, Suno, ElevenLabs and the rest.
Each entry carries the prompt format that works for that model and what it is actually good at.
◈ READ THESE AS DATED SNAPSHOTS
These were written in mid-2026. Model capabilities move faster than any guide. The prompt formats are the durable part; the capability claims carry a date and will eventually be wrong.
Check the stamp before trusting a comparison.
Production-tested guides for the tools that matter. Each card includes optimal settings, prompt format, and a copyable template.
The best generative AI models right now — with optimized prompt formats for each. Click any model to expand its full prompting guide.
AI MODELS
IMAGE GENERATION · TEXT-TO-VIDEO · AUDIO · LLM — STANDALONE AI MODELS
ChatGPT Image 2.0
OpenAI · Text-to-Image · 2026
- Selling Point: Best conversational image editing — refine through natural language in multi-turn dialogue
- Excellent text rendering directly in images
- 4-round iterative workflow produces commercial-grade output
- Understands complex multi-element scenes
+ VIEW PROMPT TEMPLATE & SAMPLE
OPTIMAL PROMPT FORMAT
Create a photorealistic portrait: 1woman, late 20s, tailored ivory blazer, golden hour backlight, Canon EOS R5 85mm f/1.2 ISO 200, shallow depth of field, Vogue Japan editorial quality, 8K.
FOLLOW-UP: "Keep everything identical but change the blazer to deep burgundy"
KEY ADVANTAGE: Use follow-up messages to iterate. "Keep everything the same but..." is the most powerful phrase in ChatGPT Image 2.0.
Nano Banana Pro 2
Speed Image Model · Rapid Prototyping · 2026
- Selling Point: Fastest image generation available — produce 10–20 concept variations in the time it takes other models to generate one
- Purpose-built for rapid iteration and prompt testing before committing to slower, higher-quality models
- Strong photorealism at speed — results good enough for social content without refinement
- Ideal workflow: prototype in Nano Banana Pro 2 → upscale and refine in ChatGPT Image 2.0 or FLUX.2 Pro Ultra
- Best for: rapid concept testing, volume social content, quick client mock-ups
+ VIEW PROMPT TEMPLATE & SAMPLE
NANO BANANA PRO 2 — OPTIMAL FORMAT
RAW photo, [SUBJECT DESCRIPTION], [LIGHTING], [CAMERA: 85mm f/1.2], [COLOR GRADE], photorealistic, ultra-detailed, 8K
Keep prompts under 60 words for fastest generation. Front-load subject and lighting. Add quality tags at the end. For iteration: run 3–5 variations, select the best, upscale with the Universal 8K Upscale prompt.
NEGATIVE: cartoon, blurry, watermark, CGI, plastic skin
SPEED WORKFLOW: Nano Banana Pro 2 → select best → Universal 8K Upscale → deliver. Full commercial-grade image in under 3 minutes.
Happy Horse
AI Image Generation · Photorealism · 2026
- Selling Point: High-fidelity photorealistic image generation with strong skin texture rendering and natural lighting reproduction
- Clean prompt adherence with detailed output — reliable for portrait, product, and lifestyle photography
- Complements the existing T2I stack — use alongside ChatGPT Image 2.0 and FLUX for a wider range of aesthetic options
- Best for: photorealistic portraits, lifestyle content, product visualization, AI influencer imagery
+ VIEW PROMPT TEMPLATE & SAMPLE
HAPPY HORSE — OPTIMAL PROMPT FORMAT
RAW photo, [SUBJECT: 1woman/1man, age, ethnicity, features], [CLOTHING DESCRIPTION], [LIGHTING: golden hour backlight / soft studio key / Rembrandt], [ENVIRONMENT: brief], shot on [CAMERA + LENS], [COLOR GRADE], photorealistic, natural skin texture, 8K ultra-detailed
SAMPLE: RAW photo, 1woman, late 20s, natural skin, black blazer, golden hour backlight from left, shallow depth of field, shot on Canon EOS R5 85mm f/1.2, warm editorial grade, photorealistic, 8K
NEGATIVE: CGI, plastic skin, airbrushed, overexposed, watermark, blurry
USE CASE: Happy Horse works well for AI influencer persona imagery and lifestyle photography. Pair with APOB AI for consistent identity across a content series.
Seedance 2.0
ByteDance · Text-to-Video · 2026
- Selling Point: Most realistic human motion in text-to-video — benchmark-leading temporal consistency in 2026
- Bracket format with [MOTION] first is mandatory for best results
- Identity lock required: "same consistent face, no morphing" prevents character drift
- Clips up to 10 seconds at 24fps — strong on fashion, lifestyle, and commercial content
- Best for: AI influencer video, cinematic human motion, brand lifestyle clips
+ VIEW PROMPT TEMPLATE & SAMPLE
BRACKET FORMAT (ALWAYS USE THIS)
[MOTION]: slow graceful walk, natural heel-to-toe rhythm
[SUBJECT]: 1woman, white dress, same face throughout, no morphing
[ENV]: Tokyo alley, midnight, neon reflections on wet cobblestones
[CAMERA]: tracking dolly, 35mm anamorphic, f/2.0
[LIGHTING]: practical neon only, warm amber and cool blue
[STYLE]: Roger Deakins cinematography, 24fps
[DURATION]: 8 seconds, no scene cuts
RULE #1: Always start with [MOTION]. This is non-negotiable for Seedance 2.0.
Kuaishou · Text-to-Video · Kling 3.0 · 2026
- Selling Point: Best physics-accurate motion — water, fire, fabric, and impact sequences look physically real
- Kling 3.0 (2026): extended clips, improved facial consistency, strong image-to-video
- Up to 10 seconds, supports slow motion and dynamic action sequences
- Best for: physics action, product reveal sequences, nature and environment video
+ VIEW PROMPT TEMPLATE & SAMPLE
OPTIMAL PROMPT FORMAT
A crystal champagne glass shatters in extreme slow motion, thousands of fragments suspended mid-air catching studio light, liquid droplets frozen in perfect spheres, black infinity background, overhead lighting creating caustic light patterns, locked camera on tripod, 135mm macro lens, 120fps, hyperrealistic glass physics, 4K
KEY TIP: Describe the physics explicitly. Name the material, the force acting on it, and the expected behavior.
Runway Gen-4
Runway · Gen-4 Turbo · 2026
- Selling Point: Best-in-class for brand commercial VFX and style transfer sequences
- Runway Gen-4 Turbo (2026): faster generation, extended clips up to 16 seconds, Act One camera control
- Strong at transformations, morphs, and surrealist visual effects for commercial use
- Best for: brand commercials, VFX sequences, style transfer, film-grade video production
+ VIEW PROMPT TEMPLATE & SAMPLE
OPTIMAL PROMPT FORMAT
A luxury perfume bottle dissolves into golden light particles that spiral upward, camera slowly orbiting, particles reconverge into a larger perfect version of the bottle against deep black infinity background, David Fincher commercial aesthetic, 4K HDR, photorealistic particle simulation, 24fps, 12 seconds, no cuts
FLUX.2 Pro
Black Forest Labs · FLUX.2 Pro Ultra · 2026
- Selling Point: Highest prompt adherence of any image model — what you write is what you get, precisely
- FLUX.2 Pro Ultra (2026): 4x resolution, faster generation, improved realism over original FLUX.2 Pro
- Optimal settings: CFG 3.5, steps 25, DPM++ 2M Karras — lower CFG than SDXL
- Best for: technical compositions, detailed architecture, precise product visualization, ComfyUI pipelines
+ VIEW PROMPT TEMPLATE & SAMPLE
OPTIMAL PROMPT FORMAT
A hyperrealistic photograph of 1woman, age 28, wearing a structured ivory blazer, standing in a rain-soaked Tokyo street at blue hour, neon reflections on wet pavement, Canon EOS R5, 85mm f/1.2, ISO 800, depth of field separation, photojournalism aesthetic, ultra-detailed skin texture, 8K resolution, no text, no watermark
Midjourney v7
Midjourney · Text-to-Image · v7 · 2026
- Selling Point: Strongest editorial aesthetic and artistic composition of any image model — distinctly recognizable visual style
- Midjourney v7 (April 2026): personalization system (--p flag), improved photorealism with --style raw, stronger style consistency
- Use --ar for aspect ratio, --no instead of negative prompts, --q 2 for quality
- Best for: editorial fashion, creative direction, concept art, artistic campaigns
+ VIEW PROMPT TEMPLATE & SAMPLE
OPTIMAL PROMPT FORMAT
editorial fashion photography, 1woman ivory blazer, golden hour backlight, Vogue Japan, 85mm f/1.2, ultra-detailed, photorealistic --ar 4:5 --style raw --v 7 --p [your-persona-id]
MJ v7 TIP: Use --p [persona-id] for consistent personal style. --style raw for max photorealism. Personalization is the biggest v7 feature — Midjourney rewards clean, intentional language.
APOB AI
AI Influencer Creation Studio · 2026
- Selling Point: Full AI persona creation with consistent face, body, and identity across unlimited content — no photography needed
- Virtual try-on, branded outfit swaps, and lifestyle scenario generation in minutes
- Ideal starting point for the full AI influencer pipeline: APOB → ChatGPT Image → Higgsfield → ElevenLabs
- Best for: AI influencer creation, brand ambassador content, fashion campaigns, e-commerce
+ VIEW PROMPT TEMPLATE & SAMPLE
HOW TO USE
1. Create your AI persona with full appearance customization
2. Generate reference sheet (front, 3/4, side views)
3. Use reference for ChatGPT Image 2.0 photo content
4. Upload to Higgsfield for video content
5. Caption with GPT-4o brand voice system prompt
NOTE: opus.pro has two tools — Agent Opus (researches a topic and builds a new video from scratch) and OpusClip Pro (takes your existing long videos and auto-cuts them into viral short clips). Both are different from Anthropic's Claude Opus model.
Beauty & Fashion AI Studio · 2026
- Selling Point: Purpose-built for beauty and fashion — virtual makeup, hair color, and outfit try-on with photorealistic skin rendering
- Sponsored post creation workflow built-in — pitch-ready brand content in one platform
- Pays recurring affiliate commissions — strong for beauty creator monetization strategy
- Best for: beauty brands, fashion influencers, product launch campaigns, e-commerce lookbooks
+ VIEW PROMPT TEMPLATE & SAMPLE
Templates: Glam AI has some genuinely great templates for beauty influencers — makeup looks, skin care routines, fashion try-ons, and sponsored beauty posts. If you're in the beauty niche, the template library alone saves hours per week. Best for: beauty brands, fashion influencers, product launch campaigns.
Udio
AI Music · Stems Export · v1.5 · 2026
- Selling Point: Stem separation and export — individual track files for post-production mixing
- Udio v1.5 (2026): improved audio fidelity, longer song generation, stronger genre coherence
- Stem separation and export — individual track files for post-production mixing
- Best for: film scoring, soundtrack production, collaborative music, stems for mixing
+ VIEW PROMPT TEMPLATE & SAMPLE
OPTIMAL PROMPT FORMAT
lo-fi hip hop, late night study session, warm vinyl crackle, dusty sample chops, muted Rhodes piano, soft brushed snare, 85 BPM, melancholic and focused, Tokyo midnight aesthetic, no vocals, 2-minute seamless loop
EXTEND: keep the same groove but introduce a subtle string arrangement in the second minute, bring Rhodes forward in the mix
SUNO vs UDIO: Suno = better for complete song production and lyrics. Udio = better for instrumental control, stems, and extending/remixing.
AI Music Generation · Suno v5.5 · 2026
- Suno v5.5 (2026) — major leap in audio quality, longer clips (up to 8 minutes), improved vocal clarity, richer instrumentation, and a full AI Music Studio workflow
- Full songs with lyrics, vocals, and production from a single text prompt — any genre, any mood
- Suno Studio (2026): multi-track editing, stem separation, and remix tools built into the platform
- Best for: AI film scores, brand content, social media, influencer soundtracks, full song production
+ VIEW PROMPT TEMPLATE & SAMPLE
SUNO V5.5 OPTIMAL PROMPT FORMAT
Genre: [cinematic electronic / indie pop / hip-hop / ambient / etc.]
Mood: [tense / uplifting / melancholic / energetic]
Instrumentation: [specific instruments + production style]
Tempo: [BPM or descriptive: slow / mid-tempo / driving]
Vocals: [male/female/none, style: raw/polished/whispered]
Energy arc: [builds from X% to Y% over Z seconds]
Reference: [Artist A meets Artist B style]
Duration: [30s / 60s / full song up to 8 minutes — v5.5]
Production quality: commercial, broadcast-ready
V5.5 UPGRADE: Suno v5.5 now supports songs up to 8 minutes with stem separation in the studio. Use "Personalize" mode to maintain a consistent sound identity across multiple tracks for a branded content series.
OpusClip Pro
AI Video Repurposing · opus.pro · 2026
- Selling Point: Upload any long-form video and it automatically finds the most viral moments, reformats for vertical, and exports short clips with captions
- AI identifies hooks, engagement peaks, and viral patterns automatically — no manual editing
- Outputs TikTok, Reels, and Shorts-ready clips with bold captions in one pass
- Best for: podcasters, YouTubers, educators, and brands turning long content into daily social posts
+ VIEW PROMPT TEMPLATE & SAMPLE
OPUSCLIP OPTIMIZATION GUIDE
For best results with OpusClip Pro:
1. Upload videos 10+ minutes long — more content = better clip variety
2. Set clip length to 45-60 seconds for TikTok/Reels
3. Enable AI captions and choose a bold font style
4. Use the "Viral Score" filter to prioritize highest-engagement clips
5. Batch export 10+ clips per video for a full week of content
WORKFLOW: Record a 30-min podcast → OpusClip Pro extracts 15 viral clips → schedule across TikTok, Reels, Shorts for 2 weeks of content automatically.
AutoClips
Faceless AI Video Automation · autoclips.app · 2026
- Selling Point: Full end-to-end faceless video automation — it writes the script, generates voiceover, creates visuals, adds captions, and auto-posts to TikTok, YouTube, Instagram, and Facebook daily.. Set it up once and it runs without you.
- Auto-posts to all 4 platforms simultaneously — with platform-optimized titles and hashtags for each
- Generates thumbnails, captions, and optimal posting times automatically — no Canva, no scheduling apps
- 40,000+ creators · Channels have reached 400K+ followers using daily AutoClips automation
- Best for: faceless YouTube channels, TikTok automation, high-volume content creators, and anyone building passive income with AI video
+ VIEW DETAILS & PRICING
AUTOCLIPS — HOW IT WORKS
1. Connect your TikTok, YouTube, Instagram, or Facebook account via secure OAuth
2. Choose your niche and video type (Character Explainer / Faceless / 3D Animation)
3. Set your posting schedule (1–3 videos per day)
4. AutoClips writes a fresh script, generates voiceover, creates visuals, adds captions, and auto-posts — every single day
AUTOMATION STACK: AutoClips for daily volume content → OpusClip Pro for clipping long-form into Shorts/Reels → ElevenLabs for custom voice consistency across all content.
↗ START AUTOMATING FREE
AI Voice & Audio Studio · 2026
- Selling Point: Most realistic AI voice synthesis available — indistinguishable from human speech at broadcast quality
- ElevenLabs AI Studio (2026): full audio production suite — voice, SFX, music, and mixing in one platform
- Voice cloning from a 30-second sample, 30+ languages, emotional tone control
- Best for: AI influencer voiceovers, podcast production, audiobooks, video narration, branded audio content
+ VIEW PROMPT TEMPLATE & SAMPLE
OPTIMAL VOICE DIRECTION PROMPT
Speak this in the voice of a confident, warm female brand ambassador in her late 20s. Pace: measured, never rushed. Tone: aspirational but approachable. Slight natural breath pauses. No robotic inflection. Emotion: excited-but-composed. Emphasis on action words.
[SCRIPT]: "The future of content is not about working harder. It is about prompting smarter."
AI STUDIO 2026: ElevenLabs now has a full production suite — generate voice, add SFX, mix music, and export broadcast-ready audio without leaving the platform. Pair with Higgsfield Studio video for fully autonomous AI influencer content.
Ready to use these models? Browse production prompts →
AI STUDIOS
CREATIVE PLATFORMS · PRODUCTION SUITES · AI AGGREGATORS
An AI Studio is an all-in-one creative platform that bundles multiple AI capabilities — image generation, video, voice, editing, and automation — into a single interface. Unlike standalone models, studios are designed for full production workflows.
AI Aggregator — platform that gives you access to multiple AI models in one place
AI Pipeline — chain of AI tools working in sequence to produce a final output
AI Persona — a consistent AI-generated character identity across content
★ There are many AI studios in the market — it comes down to where you feel comfortable and what suits your workflow. Try a few and commit to the one that fits.
Higgsfield Studio
AI Cinematic Studio · Identity Lock · 2026
- Selling Point: Not a model — a full cinematic AI production studio
- Pioneering platform for Hollywood-grade AI video production
- AI aggregator: access multiple video models in one interface
- Upload a face photo → generate that person in any scene
- Best for: AI influencers, character films, brand content
+ VIEW STUDIO GUIDE
OPTIMAL PROMPT FORMAT
[CHARACTER]: The person in the uploaded reference photo. Maintain exact facial identity throughout every frame. No morphing, no drift.
[ACTION]: walks confidently through crowd, natural relaxed stride
[CAMERA]: medium tracking shot, 35mm Cooke anamorphic, f/2.0
[ENV]: Tokyo Shibuya crossing, rush hour, evening, wet reflective pavement
[STYLE]: Ridley Scott visual language, cinematic 24fps
STUDIO TIP: Pair with APOB AI to create your AI persona, then bring them to life here.
APOB AI
AI Influencer Creation Studio · 2026
- Selling Point: Hands-on AI influencer customization — build and control every detail of your AI persona with full creative control, from appearance to identity consistency across unlimited content
- Virtual try-on, branded outfit swaps, and lifestyle scenario generation — no photography needed
- Best for: AI influencer creation, brand ambassador content, fashion and beauty campaigns
+ VIEW PROMPT TEMPLATE & SAMPLE
APOB AI PERSONA PROMPT
Create a consistent AI persona: female, late 20s, Southeast Asian, long dark hair, confident expression, athleisure style. Maintain identical face and features across all content. Generate in: outdoor lifestyle setting, golden hour, casual urban environment.
BEST WORKFLOW: Create your persona in APOB AI → generate lifestyle photos → run through ChatGPT Image 2.0 for product placement → Higgsfield for video clips.
Glam AI
Beauty & Fashion AI Studio · 2026
- Selling Point: Upload a reference image and Glam AI handles the prompting automatically — template-driven beauty and fashion content without manual prompt writing
- Sponsored post creation and brand partnership content generation built into the workflow
- Best for: beauty brands, fashion influencers, product launch campaigns, e-commerce lookbooks
+ VIEW PROMPT TEMPLATE & SAMPLE
GLAM AI CONTENT PROMPT
Beauty editorial shoot: consistent AI model, flawless skin, [MAKEUP LOOK], wearing [OUTFIT/PRODUCT], professional beauty lighting, Vogue editorial aesthetic, sponsored post format, brand-clean background.
MONETIZATION: Glam AI pays recurring affiliate commissions. Build a beauty AI influencer, pitch brands at 10K followers for $500–2,000 per sponsored post.
Research-to-Video Agent · opus.pro · 2026
- Selling Point: Give it a URL, PDF, or prompt — it researches the topic and auto-generates a complete, polished multi-scene explainer or promotional video from scratch
- No timeline editing, no assets needed — fully autonomous video production from text input
- Best for: explainer videos, thought leadership content, product demos, educational content
+ VIEW PROMPT TEMPLATE & SAMPLE
AGENT OPUS BRIEF FORMAT
Topic: [SUBJECT/URL/PDF]
Goal: Create a 60-90 second explainer video
Tone: [professional/conversational/energetic]
Key points to cover: [3-5 bullet points]
Target audience: [AUDIENCE]
Call to action: [CTA]
Style: clean, modern, branded
NOTE: Agent Opus creates NEW video from research. Different tool from OpusClip Pro (below) which edits existing video. Both are at opus.pro.
Script-to-Video Studio · 2026
- Selling Point: Converts any text, blog post, or script into a fully edited video with voiceover, music, captions, and stock footage — no editing skills needed
- 16M+ stock clips, automated scene matching, brand kit, and direct social media publishing
- Best for: marketers, YouTubers, agencies producing high-volume video content at scale
+ VIEW PROMPT TEMPLATE & SAMPLE
INVIDEO AI SCRIPT PROMPT
Create a 60-second [TOPIC] video for [PLATFORM].
Tone: [professional/casual/energetic]
Target: [AUDIENCE DESCRIPTION]
Hook: strong question or stat in first 3 seconds
Structure: hook → 3 key points → CTA
Voice: confident, clear, conversational
Music: background, not distracting
BEST FOR: Repurposing blog posts and articles into social videos. Paste a URL and InVideo AI does the rest.
Imagine Art
Multi-Model Image Studio · 2026
- Selling Point: 20+ AI image models in one platform — run the same prompt through FLUX, SDXL, Stable Diffusion variants simultaneously for comparison
- ControlNet, LoRA support, and inpainting built-in — advanced workflows without ComfyUI
- Best for: prompt engineers testing across models, designers exploring multiple visual aesthetics
+ VIEW PROMPT TEMPLATE & SAMPLE
MULTI-MODEL TEST PROMPT
RAW photo, [SUBJECT], [LIGHTING], [CAMERA + LENS], [COLOR GRADE], photorealistic, 8K, ultra-detailed
NEGATIVE: cgi, plastic, airbrushed, watermark, blurry
Run this prompt across FLUX.2 Pro, SDXL, and Playground for comparison before committing to a final model.
PRO USE: Use Imagine Art to A/B test your prompt across 3+ models before deciding which one to use for your final production run.
OpenArt AI
AI Creative Suite & Workflows · 2026
- Selling Point: Visual workflow builder lets you chain multiple AI operations together (generate → edit → upscale → stylize) without any coding
- Strong AI character consistency and avatar creation — ideal for influencer content pipelines
- Best for: creators building repeatable AI workflows, AI influencer persona development
+ VIEW PROMPT TEMPLATE & SAMPLE
OPENART CHARACTER CONSISTENCY
Consistent character portrait: [CHARACTER DESCRIPTION], same face across all generations, [STYLE: editorial/commercial/cinematic], [LIGHTING], photorealistic, high detail.
Use OpenArt's "consistent character" workflow with a reference image uploaded as the anchor for best identity lock across outputs.
WORKFLOW BUILDER: OpenArt's recipe system is the most accessible way to create repeatable AI image pipelines — no technical knowledge needed.
AI Video Studio · Free Tier · 2026
- Selling Point: Cinematic-quality text-to-video and image-to-video with a generous free tier — accessible entry point for creators new to AI video
- Strong realistic human motion and facial consistency, competitive with Kling and Runway
- Best for: cinematic short clips, AI influencer video, product showcases, social content
+ VIEW PROMPT TEMPLATE & SAMPLE
POLLO AI VIDEO PROMPT
Cinematic close-up shot, [SUBJECT] in natural environment, photorealistic skin texture, natural ambient lighting, shallow depth of field, smooth camera movement, 4K quality, no scene cuts, continuous motion throughout.
FREE TIER: Pollo AI offers daily credits — use them to prototype your AI video concepts before committing to a paid Kling or Runway subscription.
NEWSLETTER EXCLUSIVEFuture exclusive prompt packs and AI production workflows sent directly to subscribers. Subscribe free ↗