Image-to-Video Models

Overview

The Image-to-Video API provides access to 60+ models that animate static images into dynamic videos. These models automatically activate when you upload a start frame image in the Video Studio.

Image-to-video models support single or multi-image input (up to 9 images for some models) for character consistency and scene composition.

Model Categories

Multi-Image Animation Models

Advanced models that accept multiple reference images for consistent character animation.

Seedance 2.0 I2V — Up to 9 images, 5s/10s/15s duration
Openai Sora 2 / 2 Pro I2V — Multiple images with audio support
Pixverse v4.5 / v5 I2V — Multi-image with 5s/8s options
Vidu v2.0 I2V — Multiple images for 4s mobile videos
Vidu Q1 Reference — Multi-image character animation
Seedance Lite Reference Video — Multi-image reference-to-video
Wan2.1 Reference Video — Product and object animation

High-Quality Animation Models

Premium models for photorealistic and cinematic animation.

Kling v2.1 Master/Pro I2V — 5s/10s photorealistic motion
Kling v2.5/2.6 Pro I2V — Enhanced character animation
Veo3 / Veo3 Fast I2V — Google’s image-to-video models
Veo3.1 / Veo3.1 Fast I2V — 8s 1080p animation
Runway Act Two I2V — Professional cinematic effects
Leonardo AI Motion 2.0 — Slow-motion and dynamic effects

Seedance Animation Models

ByteDance’s animation models with quality and resolution controls.

Seedance Lite I2V — Fast 3–12s generation
Seedance Pro I2V — High-quality cinematic animation
Seedance 2.0 I2V — Latest with 15s support and multi-image
Seedance Lite Reference Video — Multi-image product animation

Wan Animation Models

Realistic character and portrait animation.

Wan2.2 I2V — 5s/8s portrait animation
Wan2.5 I2V / I2V Fast — Enhanced motion quality
Wan2.6 I2V — Multi-duration support (5s/10s/15s)
Wan2.1 Reference Video — Multi-image product shots

Specialized Animation Models

Domain-specific and effect-based animation tools.

Minimax Hailuo 02 Standard/Pro I2V — 6s/10s animation
OVI I2V — Creative experimental animation
Higgsfield Dop I2V — 100+ motion presets (Bullet Time, Dolly, etc.)
Video Effects — 40+ animated effects (Flying, Melt, Minecraft, etc.)

Popular Models

Model	Max Images	Duration	Resolution	Best For
Seedance 2.0 I2V	9	5s / 10s / 15s	Basic/High	Multi-image character consistency
Kling v2.6 Pro I2V	1	5s / 10s	Variable	Photorealistic single-image animation
Veo3.1 I2V	1	8s	1080p	Google-grade quality
Sora 2 Pro I2V	Multiple	10s / 15s / 25s	720p/1080p	Long-form storytelling with audio
Runway Act Two I2V	1	Variable	Variable	Professional cinematic effects
Pixverse v5 I2V	Multiple	5s / 8s	360p–1080p	Flexible multi-image animation
Wan2.6 I2V	1	5s / 10s / 15s	Variable	Portrait and character animation

Newly Added Models

These models were recently added and offer advanced animation capabilities.

Seedance 2.0 I2V

Family: ByteDance
Max Images: 9
Duration: 5s / 10s / 15s
Quality: Basic / High
Aspect Ratios: 16:9, 9:16, 4:3, 3:4

Features

Animate up to 9 reference images for character consistency
Quality presets for speed vs. fidelity tradeoff
Natural motion and physics simulation
Wide aspect ratio support
Seamless multi-image blending

Example Request

{
  "model": "seedance-v2.0-i2v",
  "images_list": [
    "https://example.com/character.jpg"
  ],
  "prompt": "The character walks forward, turns to face the camera, and waves. Smooth natural motion, cinematic lighting.",
  "aspect_ratio": "16:9",
  "duration": 10,
  "quality": "high"
}

Seedance 2.0 I2V can accept up to 9 images to maintain character consistency across complex animations.

Multi-Image Animation

Models with multi-image support maintain character appearance and style consistency across the animation.

Multi-Image Model Support

Model	Max Images	Use Case
Seedance 2.0 I2V	9	Character consistency across motion
Seedance Lite Reference Video	Multiple	Product and object animation
Openai Sora 2 / 2 Pro I2V	Multiple	Narrative with character consistency
Pixverse v4.5 / v5 I2V	Multiple	Flexible multi-reference animation
Vidu v2.0 I2V	Multiple	Mobile-optimized character animation
Vidu Q1 Reference	Multiple	Cinematic character scenes
Wan2.1 Reference Video	Multiple	Product showcase animation

Example: Multi-Image Animation Request

{
  "model": "seedance-v2.0-i2v",
  "images_list": [
    "https://example.com/character-front.jpg",
    "https://example.com/character-side.jpg",
    "https://example.com/character-back.jpg"
  ],
  "prompt": "The character walks in a circle, showcasing all angles. Smooth rotation, consistent lighting, natural movement.",
  "aspect_ratio": "16:9",
  "duration": 10,
  "quality": "high"
}

Common Input Parameters

Image URL / Images List

Type: string (single) or array (multi-image)
Required: Yes
Format: Publicly accessible HTTPS URL or Muapi uploaded file URL For single-image models, use image_url. For multi-image models, use images_list array.

Use the Upload File endpoint to host your images on Muapi infrastructure before sending them to models.

Prompt

Type: string
Required: Varies by model (optional for some motion preset models)
Max Length: 500–3000 characters Describe the desired motion, camera movement, and changes in the scene. Be specific about what should animate and how. Example Prompts:

Good: "The person waves at the camera."

Better: "The person slowly raises their right hand and waves at the camera with a gentle smile. 
Soft natural lighting, subtle head movement, camera remains static. Smooth, realistic motion."

Duration

Type: integer
Options: Varies by model (typically 4–25 seconds) Length of the animated video. Some models have fixed durations (e.g., Vidu v2.0 = 4s), while others offer flexible options.

Aspect Ratio

Type: string
Options: 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, etc. Some models auto-detect from the input image; others require explicit specification.

Resolution / Quality

Resolution: 480p, 720p, 1080p, 360p, 540p
Quality: basic, medium, high

Higher resolutions and quality settings consume more credits and take longer to generate.

Model-Specific Features

Higgsfield Dop I2V — 100+ Motion Presets

Select from curated motion and camera effects:Camera Movements:

360 Orbit, 3D Rotation, Arc Left/Right
Dolly In/Out/Left/Right, Dolly Zoom In/Out
Crane Up/Down/Over The Head
Crash Zoom In/Out, Zoom In/Out

Special Effects:

Bullet Time, Datamosh, Disintegration
Fire Breathe, Eyes In, Face Punch
Clone Explosion, Building Explosion

Actions:

Action Run, Baseball Kick, Basketball Dunks
Boxing, Catwalk, Car Chasing

{
  "model": "higgsfield-dop-image-to-video",
  "image_url": "...",
  "prompt": "Rotate the camera around the scene.",
  "motion": "360 Orbit"
}

Video Effects — 40+ Animated Effects

Apply preset animations without prompts:

Balloon Flyaway, Blow Kiss, Body Shake
Break Glass, Carry Me, Cartoon Doll
Flying, Gender Swap, Hair Swap
Minecraft, Pixel Me, Soul Depart
Zoom In Fast, Zoom Out

{
  "model": "video-effects",
  "image_url": "...",
  "name": "Flying"
}

Seedance Models — Camera Fixed Option

Camera Fixed: true | falseWhen enabled, locks the camera position and animates only the subject/scene.

{
  "model": "seedance-lite-i2v",
  "image_url": "...",
  "prompt": "The character waves at the camera.",
  "duration": 5,
  "camera_fixed": true
}

OVI I2V — Audio Captioning

Supports audio description in prompts:

Camera: static medium shot. The scientist speaks: <S>We have discovered life beyond Earth.<E> 
<AUDCAP>Soft electronic hum, distant beep of instruments<ENDAUDCAP>

Use <S>...<E> for speech and <AUDCAP>...<ENDAUDCAP> for ambient audio.

Example Use Cases

Portrait Animation

{
  "model": "wan2.6-i2v",
  "image_url": "https://example.com/portrait.jpg",
  "prompt": "The person slowly turns their head to the right, makes eye contact with the camera, and smiles gently. Soft natural lighting, subtle facial expressions, realistic motion.",
  "duration": 5
}

Product Showcase

{
  "model": "wan2.1-reference-video",
  "images_list": [
    "https://example.com/product-angle1.jpg",
    "https://example.com/product-angle2.jpg"
  ],
  "prompt": "The product rotates 360 degrees on a reflective surface, studio lighting highlighting its features.",
  "aspect_ratio": "16:9",
  "duration": 5,
  "resolution": "720p"
}

Character Consistency Animation

{
  "model": "seedance-v2.0-i2v",
  "images_list": [
    "https://example.com/char-front.jpg",
    "https://example.com/char-side.jpg"
  ],
  "prompt": "The character walks forward, stops, and looks around. Camera follows with a slow tracking shot. Natural movement, consistent lighting.",
  "aspect_ratio": "16:9",
  "duration": 10,
  "quality": "high"
}

Cinematic Effect

{
  "model": "higgsfield-dop-image-to-video",
  "image_url": "https://example.com/scene.jpg",
  "prompt": "Cinematic bullet time effect around the character.",
  "motion": "Bullet Time"
}

Duration & Resolution Guide

Use Case	Recommended Duration	Resolution	Models
Social media clips	4–8s	720p–1080p	Wan 2.6, Kling v2.6 Pro, Seedance 2.0
YouTube shorts	10–15s	1080p	Sora 2 Pro, Seedance 2.0, Wan 2.6
Product demos	5–8s	1080p	Runway Act Two, Veo 3.1, Kling v2.6 Pro
Character animation	10–15s	720p–1080p	Seedance 2.0 I2V, Sora 2 Pro
Quick effects	4–6s	720p	Video Effects, Higgsfield Dop

API Endpoints

All image-to-video models use the unified endpoint pattern:

POST /api/v1/{model-endpoint}

Refer to the Generate Image-to-Video guide for integration details.

Model Selection Tips

For character consistency: Use Seedance 2.0 I2V (9 images) or Pixverse v5 I2V (multi-image).For portrait animation: Use Wan 2.6, Kling v2.6 Pro, or Minimax Hailuo 02 Pro.For cinematic effects: Use Higgsfield Dop I2V (100+ presets) or Runway Act Two.For product animation: Use Wan2.1 Reference Video or Seedance Lite Reference.For long animations: Use Sora 2 Pro (25s) or Seedance 2.0 I2V (15s).For quick effects: Use Video Effects or Higgsfield Dop with motion presets.

Best Practices

Upload First: Use the Upload File endpoint to host images before animating.
Start Frame Quality: Use high-quality, well-lit images for best results.
Be Specific: Describe motion, camera movement, and what should remain static.
Test Duration: Start with 5s generations before committing to longer durations.
Camera Fixed: Use camera_fixed: true for subject-only animation without camera movement.
Multi-Image Order: For multi-image models, sequence matters—order images logically.
Resolution Trade-offs: Use 720p for iteration, 1080p for final output.

Rate Limits

Image-to-video generation consumes more credits than text-to-video through the Muapi.ai API. Cost factors include:

Duration: Longer videos cost more
Resolution: Higher resolutions cost more
Quality: High-quality modes cost more
Multi-Image: Multiple input images may cost more
Model: Premium models cost more than open-source options

Refer to the Muapi.ai documentation for detailed pricing information.

API Client

Models

Overview

Model Categories

Popular Models

Newly Added Models

Seedance 2.0 I2V

Multi-Image Animation

Multi-Image Model Support

Example: Multi-Image Animation Request

Common Input Parameters

Image URL / Images List

Prompt

Duration

Aspect Ratio

Resolution / Quality

Model-Specific Features

Example Use Cases

Duration & Resolution Guide

API Endpoints

Model Selection Tips

Best Practices

Rate Limits

Build docs developers (and LLMs) love

API Client

Models

​Overview

​Model Categories

​Popular Models

​Newly Added Models

​Seedance 2.0 I2V

​Multi-Image Animation

​Multi-Image Model Support

​Example: Multi-Image Animation Request

​Common Input Parameters

​Image URL / Images List

​Prompt

​Duration

​Aspect Ratio

​Resolution / Quality

​Model-Specific Features

​Example Use Cases

​Duration & Resolution Guide

​API Endpoints

​Model Selection Tips

​Best Practices

​Rate Limits

Build docs developers (and LLMs) love

Overview

Model Categories

Popular Models

Newly Added Models

Seedance 2.0 I2V

Multi-Image Animation

Multi-Image Model Support

Example: Multi-Image Animation Request

Common Input Parameters

Image URL / Images List

Prompt

Duration

Aspect Ratio

Resolution / Quality

Model-Specific Features

Example Use Cases

Duration & Resolution Guide

API Endpoints

Model Selection Tips

Best Practices

Rate Limits