muapi Logomuapi Logo
muapi Logomuapi Logo

Seedance 2 Mini Omni Reference: AI Video Generator

Generate videos with image, video, and audio references using Seedance 2 Mini. Affordable omni-mode at 480p–720p. Try free — pay per generation. Seedance 2 Mini Omni Reference generates video from a text prompt with optional image, video, and audio references. Cost-efficient mini-tier model for reference-driven workflows.

Interactive model controls

Seedance 2 Mini Omni Reference generates video from a text prompt with optional image, video, and audio references. Cost-efficient mini-tier model for reference-driven workflows.

📝

Overview

About this model

Seedance 2 Mini Omni Reference is the most flexible mini-tier generation mode, accepting up to 9 reference images, 3 reference video clips, and 3 reference audio files alongside your text prompt. Reference each asset in your prompt with @image1–@image9, @video1–@video3, and @audio1–@audio3 for precise creative control.

1Character consistency: Keep a character's appearance consistent across multiple shots using reference images.
2Motion reference: Supply a short video clip to guide the motion style of the generated video.
3Audio-driven: Sync generated visuals to a reference audio file such as music or speech.
4Multi-modal storytelling: Combine image, video, and audio references for complex creative workflows.
💰

Pricing & Value

Cost analysis

muapiapp$0.15/sec at 720p + video surcharge, $0.08/sec at 480p + video surcharge

Video reference surcharge: 30% of rate × total input video seconds.

Fal.aiNot available

Seedance 2 Mini Omni Reference not available via Fal.ai.

ReplicateNot available

Seedance 2 Mini Omni Reference not available via Replicate.

** Competitor pricing is estimated based on similar model architectures and usage tiers.

⚙️

Technical Details

Configuration schema

Promptstring

Text prompt. Reference images with @image1..@image9, videos with @video1..@video3, audio with @audio1..@audio3.

Default ValueThe character walks forward confidently in a sunny meadow, camera follows from behind.
Reference Imagesarray

Up to 9 reference images (JPEG/PNG/WebP). Referenced in prompt via @image1..@image9.

Default Valuehttps://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/seedance-v1.5-pro-i2v.jpg
Reference Videosarray

Up to 3 reference video clips (MP4, total max 15s). Referenced in prompt via @video1..@video3.

Default Value
Reference Audioarray

Up to 3 reference audio files (MP3/WAV, total max 15s). Referenced in prompt via @audio1..@audio3.

Default Value
Aspect RatioEnum (6 options)

Aspect ratio of the output video.

Default Value16:9
Durationint

Video duration in seconds.

Default Value5
ResolutionEnum (2 options)

Output video resolution.

Default Value720p
Generate Audioboolean

Whether to generate AI audio synchronized with the video.

Default Valuetrue
High Bitrateboolean

Enable high bitrate mode for better visual fidelity. Produces larger files.

Default Valuefalse
📖

Implementation Guide

Developer documentation

How to Use Seedance 2 Mini Omni Reference

  1. Write your prompt: Describe the scene and reference any uploaded assets with @image1, @video1, @audio1, etc.

  2. Upload reference assets (all optional):

    • images_list: Up to 9 reference images (@image1–@image9)
    • video_files: Up to 3 video clips, total max 15s (@video1–@video3)
    • audio_files: Up to 3 audio files, total max 15s (@audio1–@audio3)

  3. Set resolution, duration, and aspect ratio.

  4. High bitrate (optional): For better visual fidelity.

  5. Submit and poll /api/v1/predictions/{id}/result for results.

API Example

curl -X POST https://api.muapi.ai/api/v1/seedance-2-mini-omni-reference \
  -H 'x-api-key: YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
    "prompt": "@image1 is the character. The person walks confidently in a sunny meadow",
    "images_list": ["https://example.com/character.jpg"],
    "aspect_ratio": "16:9",
    "duration": 5,
    "resolution": "720p"
  }'
❓

Common Questions

Frequently asked

What is the video reference duration limit?

The total duration of all reference video clips must not exceed 15 seconds. Each clip should be an MP4 file.

Does using video references change the cost?

Yes. The cost is base rate × output duration, plus 30% of rate × total input video duration for each second of reference video provided.

How do I reference uploaded assets in the prompt?

Use @image1 through @image9 for images, @video1 through @video3 for videos, and @audio1 through @audio3 for audio files, in the order they were provided in the respective arrays.

What happens if I don't provide any reference assets?

If no images, videos, or audio are provided, the model behaves like text-to-video and generates a clip from the prompt alone.

Can I mix images and videos in the same request?

Yes. You can simultaneously provide images_list, video_files, and audio_files. Reference each in your prompt with @image1, @video1, @audio1 etc.