How to Create Motion Prompts

Properly setting your motion prompts brings your Avatar IV creations to life by giving you full control over body movement, gestures, and expression. Doing this allows you to describe exactly how your avatar should move or react within a scene, helping you create natural, dynamic results every time.

Getting Started

When using Avatar IV, you can direct your avatar’s motion using short text prompts.

In the AI Studio, click your avatar, select the person icon in the upper-right corner, and choose your avatar engine.

You can choose from several available engines: Avatar IV, Avatar Unlimited (formerly Avatar III), Kling, Seedance, Runway, and Hailuo.

Avatar IV – Generates realistic, expressive facial and upper-body movements for high-quality, natural-looking avatar performance.
Avatar Unlimited (formerly Avatar III) – Offers unlimited usage with fast, reliable lip-sync only—ideal for bulk production and automated workflows.
Kling – Produces high-fidelity, cinematic motion suited for storytelling and polished, visually rich content.
Seedance – Prioritizes speed and efficiency, making it ideal for rapid iterations or high-volume video creation.
Runway – Delivers stable, consistent motion generation designed for dependable, professional-grade workflows.
Hailuo – Provides advanced, detailed motion output for complex scenes and more expressive animation requirements.

Writing Motion Prompts with Avatar IV

Each Motion Prompt describes what the avatar should do and how it should do it.

Use the following simple structure:
[Body part] + [Action] + [Emotion or intensity]

Example:

Right arm raises to wave enthusiastically.

Keep your descriptions short and clear. Each line should represent a single movement idea. Avoid combining multiple actions in one sentence — separate them for better results.

Best Practices for Writing Motion Prompts

Be specific. Focus on clear, visible actions: point, nod, turn, wave, smile.
Avoid overloading. Limit each prompt to one gesture or emotion at a time.
Keep timing natural. Avatar IV automatically handles pacing — there’s no need to specify seconds.
Control facial expressions. You can include emotions directly in the prompt (for example, smiles gently, looks surprised).
Experiment. Small wording changes can produce big differences in performance.

Using Preset Prompts

HeyGen provides built-in preset motion prompts that you can use immediately. These presets are longer, carefully designed motion sequences that look natural across different scenes.

You can:

Use a preset as-is for quick results
Edit or shorten it to match your own tone or context

Preset prompts are a great starting point if you’re new to motion design.

Motion Prompt Structure Examples

Type	Example Prompt
Simple Gesture	Avatar waves with a friendly smile.
Emotional Expression	Avatar crosses arms, looking thoughtful.
Multi-Part Sequence	Avatar nods confidently, then points forward.

Keep prompts under two short clauses for best stability and lifelike motion.

FAQ

Q: Can I use Motion Prompts with any avatar?
Yes. Motion Prompts work with all Avatar IV avatars.

Q: Do I need to specify timing or speed?
No. The system automatically controls timing and pacing.

Q: Can I combine gestures and facial expressions?
Yes, just keep them concise. Example: Avatar smiles softly while raising a hand.

Writing Motion Prompts with Veo 3.1

HeyGen x VEO 3.1 lets you create full cinematic avatar videos directly in-platform. You can design complete scenes with camera motion, backgrounds, and multi-speaker support — built scene-by-scene.

Each scene lasts up to 8 seconds, and you can add up to 8 scenes per video.
Keep prompts short, concrete, and scene-specific.

Avatar IV vs. VEO 3.1

Feature	Avatar IV	VEO 3.1
Body & face control	✅ Precise gesture & expression control	✅ Includes all Avatar IV features
Cinematic background	—	✅ Auto-generated sets, props, lighting
Camera motion	—	✅ Pans, zooms, and tilts
Multi-speaker sequencing	—	✅ Supports dialogue between characters
Scene duration	180 seconds (3 minutes) per scene	8-second scenes, up to 8 per video

Use Avatar IV when you want tight performance detail (hand movement, eye motion).
Use VEO 3.1 when you want your avatar in a cinematic context with camera movement and environmental storytelling.

How to Write a Reliable VEO 3.1 Scene Prompt

Keep your prompt short, structured, and limited to essential details.
Use this simple format:

Intent – What happens in this scene?
Avatar instruction – Body + action + emotion.
Camera / background / effects – Include lighting, FX, and movement.
Dialogue/script – The line(s) spoken in this scene.

Best Practices

Each scene = 8 seconds, so plan actions that fit that time window.
Use strong verbs (jump, point, turn, wave, land).
Keep lines atomic — 1–2 short clauses only.
Call out FX and assets explicitly (add electric spark FX on landing; use electric_spark.mp4 overlay).
Reference assets by exact name or URL, not temporary blob: links.
Test one scene first, keep voice engine and locale consistent, and run in Chrome (desktop) for best performance.
Remember: AI is not perfectly deterministic — re-running the same prompt a few times can refine results.

Example Scene Prompt

Intent: Hero greets the audience, then shows agility.
Avatar: Says “Hey, guys!”, jumps with an explosive backflip, lands in a low heroic stance facing the green villain; confident and playful.
Camera / background / effects: Wide LA skyline in daylight with a flying spaceship; follow the flip with a quick pan and slight tilt; add electric-spark FX on landing (electric_spark.mp4).
Dialogue: “Hey, guys!”

Tips for Cinematic Motion

Describe the camera movement: use verbs like “slow pan,” “smooth dolly-in,” “handheld tracking,” or “locked camera” instead of saying what not to do.
Set the mood and lighting: add cues such as “golden-hour sunlight,” “cold blue moonlight,” or “soft studio glow.”
Add physicality: mention small gestures or posture shifts (“subtle nod,” “leans forward,” “turns slightly toward light”) for realistic motion.
Use cinematic framing: specify shot types (“wide establishing,” “medium close-up,” “over-the-shoulder”) and depth of field for focus control.
Give emotional rhythm: write prompts that suggest pacing or tone (“slow, contemplative,” “dynamic, fast-cut,” “dramatic tension builds”).