Unlock Wan 2.2 Hidden Powers: 200+ Viral Prompts That Make Videos Go BOOM
Table of Contents

This comprehensive prompt library provides ready-to-use templates for Wan 2.2, leveraging its enhanced Mixture-of-Experts (MoE) architecture and cinematic-level aesthetic control. Each prompt has been tested and optimized for maximum quality and consistency.
Cinematic Shots (Text-to-Video)
Professional Camera Movements
Scene Type | Ready-to-Copy Prompt | Why It Works |
---|
Cyberpunk night-market tracking | A rainy night in a neon-soaked cyberpunk street market. Shoulder-height camera steadily follows a hooded courier weaving through crowds beneath flickering kanji signs. Pink-blue volumetric back-light, puddles reflecting neon, shallow depth of field, Blade-Runner mood. | Combines environment, camera movement, lighting and mood keywords for maximal continuity. |
Alpine reveal dolly-tilt | Extreme close-up of an ice axe striking rock. Camera dollies backward and tilts up to reveal a lone mountaineer on a sunrise-lit alpine ridge. Crisp morning air, golden rim-light, subtle lens flare. | Specifies two simultaneous moves (dolly + tilt) and clear time-of-day cues. |
Slow-motion orca 360° | An orca breaches in crystal-clear Arctic water. Ultra-slow-motion 360-degree orbit around the whale; suspended droplets sparkle in pastel polar sunset. Cinematic HDR. | Demonstrates large-scale subject motion plus rotational camera path. |
Establishing Shots
Scene Type | Prompt | Best Resolution |
---|
Epic landscape opener | An epic aerial shot of snow-capped mountains at sunrise, golden light casting long shadows, shot on 35mm, 24fps, wide-angle lens, cinematic grade | 1280×720 @ 24fps |
Urban dystopia fly-through | A neon-lit cyberpunk megacity at night, drone perspective weaving between skyscrapers, reflective rain-slick streets, 24fps, 720P | 1280×720 @ 24fps |
Pastoral countryside pan | Golden wheat field swaying in gentle breeze, camera slowly pans left to reveal distant farmhouse, warm afternoon light, peaceful rural atmosphere | 960×540 @ 16fps (draft) |
Character & Creature Moments
Anthropomorphic Characters
Scene Description | Prompt | Recommended Input Mode |
---|
Boxing-ring cats | Two anthropomorphic cats in vintage boxing trunks and bright gloves trade punches under a single spotlight. Ringside crowd fades into darkness, cinematic dust particles float. | Pure T2V or TI2V with a still reference frame for style lock-in. |
Sunglass-surfing cat | White cat wearing aviator sunglasses lounges on a pastel surfboard, gentle summer beach waves rolling behind, dreamy bokeh, 1980s postcard aesthetic. | I2V/TI2V – drop in a starting illustration of the cat for identity consistency. |
Human Portraits & Emotions
Emotion/Action | Prompt Template | Technical Notes |
---|
Subtle smile reveal | Portrait of [subject], slowly looks up and smiles warmly at camera, soft natural lighting, shallow depth of field, genuine expression | Use I2V with high-quality portrait photo for best results |
Contemplative moment | [Subject] gazing thoughtfully out window, gentle breeze moving hair, golden hour lighting, cinematic mood, peaceful atmosphere | Works well with TI2V-5B hybrid model |
Joyful laughter | [Subject] throws head back in genuine laughter, eyes crinkled with joy, warm studio lighting, natural movement, authentic emotion | Best with 720P resolution for facial detail |
Action & Emotion Short Prompts
Useful when you need rapid iterations or fine-tune timing; keep each to ≤10 words.
Dance & Movement
Girl dances slow-motion street style at night
Ballet dancer spins gracefully in spotlight
Breakdancer freezes mid-move, dramatic lighting
Emotional Expressions
Young woman lifts head and smiles softly
She turns away in tears, rain falling
Child's eyes widen with wonder and excitement
Nature & Wildlife
Eagle soars majestically over mountain peaks
Butterfly lands delicately on flower petal
Ocean waves crash against rocky coastline
Negative Prompt Starter Pack
Universal Negative Prompts
bright colors, overexposed, static, blurred details, subtitles,
worst quality, jpeg artifacts, ugly, extra fingers, poorly drawn
hands, still image, cluttered background, duplicate limbs, random text
Specific Use Case Negatives
Use Case | Additional Negative Prompts |
---|
Portrait/Character | deformed face, asymmetrical eyes, bad anatomy, multiple heads, extra limbs |
Landscape/Nature | urban elements, buildings, people, text overlays, watermarks |
Product Showcase | cluttered background, distracting elements, poor lighting, shadows on product |
Action Sequences | motion blur, static pose, frozen movement, choppy animation |
💡 Pro Tip: Wan 2.2 obeys negative prompts aggressively. Start with the universal list above every time, then append task-specific bans (e.g., "no crowds" or "no watermarks") as needed.
Practical Production Tips
1. Optimal Prompt Length
- English: 80-120 words delivers the richest results
- Chinese: Aim for 50-80 characters if working in Chinese
- Key Elements: Include camera angle, lighting, mood, and movement
2. Resolution & FPS Guidelines
Purpose | Resolution | FPS | Model Recommendation |
---|
Draft/Preview | 960×540 | 16fps | TI2V-5B (8GB VRAM) |
Final/Publishing | 1280×720 | 24fps | T2V-A14B (16GB+ VRAM) |
High-Quality | 1280×720 | 24fps | I2V-A14B (16GB+ VRAM) |
3. Hybrid Input (TI2V-5B)
- Include both text and a reference image to lock character likeness or overall palette
- Best for maintaining consistency across multiple generations
- Ideal for brand content or character-driven narratives
4. Prompt Auto-Expansion
Use the official generate.py --use_prompt_extend
flag (DashScope or local Qwen) to automatically enrich:
- Camera directions and movements
- Weather and atmospheric cues
- Micro-actions and details
- Cinematic terminology
5. Iteration Workflow
Step-by-Step Process:
- Generate low-res preview with seed fixed (960×540 @ 16fps)
- Adjust wording or negative list if artifacts appear
- Test variations with different seed values
- Upscale or re-render at target resolution only after visual approval
- Apply post-processing if needed for final delivery
Advanced Techniques
Camera Language for Wan 2.2
Camera Movement | Prompt Keywords | Example Usage |
---|
Dolly | camera dollies in/out , push-in , pull-back | Camera dollies slowly toward the subject |
Pan | pans left/right , sweeping pan , gentle pan | Camera pans across the landscape |
Tilt | tilts up/down , low angle tilt up | Camera tilts up to reveal the sky |
Orbit | 360-degree orbit , circular movement around | Camera orbits around the dancing figure |
Tracking | follows , tracks alongside , steadicam follow | Camera tracks the running athlete |
Lighting Terminology
Lighting Type | Keywords | Mood Effect |
---|
Golden Hour | golden light , warm sunset , rim lighting | Romantic, cinematic |
Blue Hour | twilight , blue hour , cool evening light | Mysterious, atmospheric |
Studio | soft box lighting , key light , rim light | Professional, controlled |
Natural | window light , overcast , natural illumination | Authentic, realistic |
Dramatic | single spotlight , high contrast , chiaroscuro | Intense, artistic |
Motion Description Best Practices
- Be specific: "gentle sway" vs "movement"
- Include timing: "slow motion", "real-time", "accelerated"
- Describe physics: "gravity", "momentum", "weightless"
- Use cinematic terms: "whip pan", "crash zoom", "rack focus"
Contributing
How to Expand This Library
Have you discovered a better Wan 2.2 prompt? Here's how to contribute:
- Test thoroughly - Ensure your prompt works consistently across multiple generations
- Document parameters - Include model variant, resolution, and seed if applicable
- Explain effectiveness - Note why specific word choices improve results
- Share variations - Provide alternatives for different styles or moods
Community Resources
This prompt library is continuously updated based on community feedback and Wan 2.2 model improvements. For the latest techniques and best practices, visit the WanVideo Official Site.