Video Studio

The Video Studio is a dual-mode AI video generation workspace that intelligently switches between text-to-video (T2V) and image-to-video (I2V) workflows based on whether you’ve uploaded a start frame.

Dual Mode Intelligence

Video Studio features the same smart mode switching as Image Studio, adapting the entire interface to your workflow:

Text-to-Video Mode

Generate videos from text descriptions. Access 40+ models including Kling, Sora, Veo, Wan, Seedance 2.0, and Runway.

Image-to-Video Mode

Animate images into videos. Upload a start frame to switch to 60+ I2V models that bring still images to life.

How Mode Switching Works

Start in T2V Mode — Default to text-to-video with 40+ creative models
Upload triggers I2V Mode — Upload an image to automatically switch to 60+ image-to-video models
Clear returns to T2V — Remove the start frame to return to text-to-video mode

The model selector, aspect ratio, duration, and quality controls all update automatically when you switch modes.

Generating Videos from Text

Enter Your Prompt

Describe the video scene you want to create. The prompt is required for text-to-video generation.

Example: "A time-lapse of a city skyline transitioning from day to night, with lights turning on in buildings"

Select Your Model

Click the model button to browse 40+ text-to-video models:Flagship Models:

Seedance 2.0 — ByteDance · 16:9/9:16/4:3/3:4 · 5/10/15s · Quality basic/high
Kling v3 — Kuaishou’s latest · High motion quality
Sora 2 — OpenAI’s video generation · Natural motion
Veo 3 — Google’s video model · Smooth dynamics
Wan 2.6 — Chinese model · Quality presets
Runway Gen-3 — Professional video synthesis
Hailuo 2.3 — Fast generation · Good quality

Use the search bar to quickly find models by name.

Configure Video Settings

Adjust the video parameters (available controls vary by model):Aspect Ratio:

16:9 — Landscape (YouTube, desktop)
9:16 — Portrait (TikTok, Instagram Reels)
4:3, 3:4 — Traditional and portrait formats
1:1 — Square (Instagram feed)

Duration:

5s — Quick clips
10s — Standard length
15s — Extended scenes (if supported)

Quality/Resolution (if available):

basic / high quality for Seedance 2.0
Resolution presets for other models

Generate Video

Click Generate ✨ to start. Video generation typically takes:

5-10 seconds: Fast models (Hailuo, basic quality)
30-60 seconds: Standard models (Kling, Wan)
1-3 minutes: Premium models (Sora, Veo)

The interface shows a loading spinner and polls the API until completion.

Animating Images to Video

Upload a Start Frame

Click the upload button (📤 icon) to open the reference image picker:

Select from upload history — Reuse previously uploaded images
Upload a new image — Choose a file from your device

The studio automatically switches to I2V mode when you select an image.

Studio Switches to I2V Mode

When you upload a start frame, the studio automatically:

Switches to 60+ image-to-video models
Updates the prompt placeholder to “Describe the motion or effect (optional)”
Shows duration and quality controls for compatible I2V models

Enter Motion Prompt (Optional)

For image-to-video generation, the prompt is optional. Many models can animate images without text:

Example: "Slow zoom in with camera rotating clockwise"
Example: "Gentle wind blowing through hair and clothes"
Example: "Water rippling and splashing"

Leave the prompt empty to let the model decide the animation.

Select I2V Model

Browse 60+ image-to-video models:

Seedance 2.0 I2V — ByteDance · Up to 9 reference images · 5/10/15s · Quality basic/high
Kling v2.1 I2V — High-quality image animation
Veo3 I2V — Google’s image-to-video model
Runway I2V — Professional animation
Wan I2V — Quality presets available
Midjourney v7 I2V — Creative animation styles

Generate Animation

Click Generate ✨ to animate your image. The API will:

Upload your start frame (if not already uploaded)
Send the image URL and optional motion prompt to the model
Return the animated video

Video Tools (V2V Mode)

Video Studio also includes video-to-video tools for processing existing videos:

Accessing V2V Tools

Click the video upload button (🎥 icon next to the image upload button)
Select a video file from your device
The studio switches to V2V mode and shows video processing models

Available Tools

Watermark Removal — Clean removal of video watermarks
Video Upscaling — Enhance resolution and quality
Video Restoration — Improve old or degraded videos

In V2V mode, parameter controls (aspect ratio, duration, etc.) are hidden because the output matches the input video format.

Supported Models

Text-to-Video Models (40+)

Flagship Models
Additional Models

Model	Key Features
Seedance 2.0	ByteDance · Aspect ratios 16:9/9:16/4:3/3:4 · Duration 5/10/15s · Quality basic/high
Kling v3	Kuaishou · High motion quality · Natural physics
Sora 2	OpenAI · Natural motion · Advanced composition
Veo 3	Google · Smooth dynamics · Cinematic quality
Wan 2.6	Chinese model · Quality presets · Fast generation
Runway Gen-3	Professional synthesis · Artistic control

Image-to-Video Models (60+)

Animation Models
Additional I2V Models

Model	Max Images	Key Features
Seedance 2.0 I2V	9	ByteDance · 16:9/9:16/4:3/3:4 · 5/10/15s · Quality basic/high
Kling v2.1 I2V	1	High-quality animation · Natural motion
Veo3 I2V	1	Google · Smooth animation
Runway I2V	1	Professional quality · Artistic styles
Midjourney v7 I2V	1	Creative animation · Unique styles

Advanced Features

Video Extension (Seedance 2.0)

When you generate a video using Seedance 2.0 T2V or Seedance 2.0 I2V, an ↗ Extend button appears in the canvas controls.

Generate Initial Video

Create a video using Seedance 2.0 (T2V or I2V mode).

Click Extend Button

After the video completes, click ↗ Extend in the canvas controls.

Studio Switches to Extend Mode

The interface returns to the prompt bar and automatically:

Switches to Seedance 2.0 Extend model
Shows an extend banner: “Extending previous Seedance 2.0 generation”
Stores the request_id from your previous generation

Add Continuation Prompt (Optional)

Enter an optional prompt to guide how the video continues:

Example: "Camera continues panning to reveal a mountain range"

Generate Extension

Click Generate ✨ to seamlessly extend your video. The extension will:

Preserve style, motion, and audio from the original
Continue for 5/10/15s (configurable)
Maintain visual consistency

The Extend feature requires a valid request_id from a Seedance 2.0 generation. If you switch models or clear history, the extend context is lost.

Duration and Aspect Ratio Controls

Video Studio’s control bar adapts to show only options supported by the selected model: Aspect Ratio Button:

Shows available aspect ratios (16:9, 9:16, 4:3, 3:4, 1:1, 21:9)
Hidden if the model doesn’t support aspect ratio selection

Duration Button:

Shows available durations (5s, 10s, 15s, 20s, etc.)
Hidden if the model uses fixed duration

Quality Button:

Shows quality presets (basic, high) or resolution options
Hidden if the model doesn’t support quality selection

Controls automatically appear/disappear when you switch models, keeping the interface clean and focused.

Canvas & Video Playback

When your video generation completes, the Video Studio transitions to the canvas view:

Video Player

Your video plays automatically in a native HTML5 video player with controls, looping, and autoplay (muted).

Action Buttons

↻ Regenerate — Generate the same video again
↗ Extend — Continue the video (Seedance 2.0 only)
↓ Download — Save video as MP4
+ New — Return to prompt bar for a new video

Video History Sidebar

All video generations are saved to a history sidebar:

Persistent storage — Up to 30 videos saved in browser localStorage
Video thumbnail previews — Quick visual reference with hover preview
Click to load — Click any thumbnail to play in canvas
Download from history — Hover to download past videos

Video history is stored separately from image history. Each studio maintains its own generation history.

Model Selection Tips

Best for Natural Motion

Sora 2 — Most realistic physics and natural movement
Veo 3 — Smooth, cinematic motion
Kling v3 — High-quality motion with good physics

Best for Speed

Hailuo 2.3 — Fastest generation, good quality
Seedance 2.0 (basic quality) — Fast with decent results
Wan 2.6 — Quick generation with quality presets

Best for Quality

Sora 2 — Highest overall quality
Veo 3 — Cinematic quality, professional results
Seedance 2.0 (high quality) — Best ByteDance quality
Runway Gen-3 — Professional synthesis

Best for Animation

Midjourney v7 I2V — Creative styles, artistic animation
Runway I2V — Professional image animation
Seedance 2.0 I2V — Multi-image support (up to 9 images)
Veo3 I2V — Smooth, natural animation

Tips for Best Results

Writing Effective Video Prompts

Describe motion — “Camera pans left”, “zoom in slowly”, “object rotates”
Set the scene — Time of day, location, weather, atmosphere
Include action — What happens in the video from start to finish
Add style — “cinematic”, “time-lapse”, “slow motion”, “aerial view”
Keep it focused — 5-10s videos work best with simple, focused scenes

Image-to-Video Best Practices

Use high-quality start frames — Clear, well-composed images work best
Choose images with potential motion — Hair, water, clouds, fabric
Avoid still portraits — Models struggle to animate completely still subjects
Add motion prompts — Guide the animation with specific motion descriptions
Try different models — Each model animates images differently

Duration and Quality Trade-offs

5s videos — Fastest generation, most reliable
10s videos — Good balance of length and quality
15s+ videos — Slower generation, may have consistency issues
Basic quality — 2-3x faster, good for previews
High quality — Slower, better detail and consistency

API Integration

Video Studio uses the Muapi.ai unified API:

// Text-to-Video generation
const result = await muapi.generateVideo({
  model: 'seedance-v2.0-t2v',
  prompt: 'A cinematic shot of a city at sunset',
  aspect_ratio: '16:9',
  duration: 10,
  quality: 'high'
});

// Image-to-Video generation
const result = await muapi.generateI2V({
  model: 'seedance-v2.0-i2v',
  image_url: 'https://...',
  prompt: 'Gentle camera movement',
  aspect_ratio: '16:9',
  duration: 5,
  quality: 'basic'
});

// Video extension
const result = await muapi.generateVideo({
  model: 'seedance-v2.0-extend',
  request_id: 'previous_generation_id',
  prompt: 'Continue the scene',
  duration: 5
});

See the API Reference for complete documentation.

Upload History

Reuse previously uploaded start frames

Generation History

Browse and download past video generations

Cinema Studio

Professional camera controls for cinematic shots

Get Started

Core Features

Guides

Dual Mode Intelligence

Text-to-Video Mode

Image-to-Video Mode

How Mode Switching Works

Generating Videos from Text

Animating Images to Video

Video Tools (V2V Mode)

Accessing V2V Tools

Available Tools

Supported Models

Text-to-Video Models (40+)

Image-to-Video Models (60+)

Advanced Features

Video Extension (Seedance 2.0)

Duration and Aspect Ratio Controls

Canvas & Video Playback

Video Player

Action Buttons

Video History Sidebar

Model Selection Tips

Tips for Best Results

API Integration

Upload History

Generation History

Cinema Studio

Build docs developers (and LLMs) love

Get Started

Core Features

Guides

​Dual Mode Intelligence

Text-to-Video Mode

Image-to-Video Mode

​How Mode Switching Works

​Generating Videos from Text

​Animating Images to Video

​Video Tools (V2V Mode)

​Accessing V2V Tools

​Available Tools

​Supported Models

​Text-to-Video Models (40+)

​Image-to-Video Models (60+)

​Advanced Features

​Video Extension (Seedance 2.0)

​Duration and Aspect Ratio Controls

​Canvas & Video Playback

Video Player

Action Buttons

​Video History Sidebar

​Model Selection Tips

​Tips for Best Results

​API Integration

​Related Features

Upload History

Generation History

Cinema Studio

Build docs developers (and LLMs) love

Dual Mode Intelligence

How Mode Switching Works

Generating Videos from Text

Animating Images to Video

Video Tools (V2V Mode)

Accessing V2V Tools

Available Tools

Supported Models

Text-to-Video Models (40+)

Image-to-Video Models (60+)

Advanced Features

Video Extension (Seedance 2.0)

Duration and Aspect Ratio Controls

Canvas & Video Playback

Video History Sidebar

Model Selection Tips

Tips for Best Results

API Integration

Related Features