Omni Reference VIP 4K: AI Image-to-Video Generator

SD 2 Omni Reference VIP 4K by ByteDance. Generate 4K ultra-HD videos using up to 9 image references, up to 3 video clips, and up to 3 audio references with priority routing. Reference materials in your prompt with @image1…@image9, @video1…@video3, and @audio1…@audio3.

📝

Overview

About this model

Updated Jul 16, 2026

Seedance 2 Omni Reference VIP 4K extends our multi-modal reference system to 4K ultra-HD quality. Combine up to 9 images, 3 video clips, and 3 audio files into a single 4K generation using the same intuitive reference notation (@image1-@image9, @video1-@video3, @audio1-@audio3). The model synthesizes all these materials into a pristine 3840×2160 video with full audio-visual synchronization, generating 4 to 15 second outputs that respect the visual language and audio character of every reference you provide.

Ideal for premium production work where quality, complexity, and creative ambition converge. Professional studios, high-end agencies, and ambitious indie creators use this tier to produce broadcast-grade content that blends multiple sources into seamless, cohesive videos. Muapi's unified API and per-generation pricing let you create at any scale without subscription commitments.

1Produce broadcast-quality commercials that weave together product shots, voiceovers, and stock footage

2Create cinematic trailers by synthesizing mood boards, existing footage, and narration tracks

3Generate high-end music videos combining multiple visual style references and professional audio

4Deliver 4K edited content to clients that respects multiple source materials in a single generation

💰

Pricing & Value

Cost analysis

Provider	Cost	Notes
muapiapp	$6.75 per generation	Muapi's transparent per-generation model ensures predictable costs for complex multi-modal 4K productions, with no subscription overhead or hidden reference-complexity surcharges.

muapiapp$6.75 per generation

Muapi's transparent per-generation model ensures predictable costs for complex multi-modal 4K productions, with no subscription overhead or hidden reference-complexity surcharges.

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

⚙️

Technical Details

Configuration schema

Parameter	Type	Description	Default
Prompt	string	Video description. Use @image1…@image9 to reference images, @video1…@video3 for videos, and @audio1…@audio3 for audio. Use @character:<request_id> for a Seedance 2 character sheet or @omni-character:<char_id> for a trained Kinovi character. Multiple characters are supported.	`@image1 is the main character. The person walks along a city street at sunset, cinematic lighting.`
Image URLs	array	Up to 9 reference image URLs (JPEG/PNG/WebP). Each Nth image corresponds to @imageN in the prompt.	`https://d3adwkbyhxyrtq.cloudfront.net/ai-images/186/712345784292/4a8c5c70-abcc-4920-873e-b0e219986453.jpg`
Video Reference URLs	array	Up to 3 reference video clip URLs (MP4, max 15s each). Each Nth video corresponds to @videoN in the prompt.	`undefined`
Audio Reference URLs	array	Up to 3 reference audio files (MP3/WAV, total max 15s). Each Nth audio corresponds to @audioN in the prompt.	`undefined`
Aspect Ratio	Enum (6 options)	Output video aspect ratio.	`16:9`
Duration (seconds)	int	Video duration in seconds.	`5`

Promptstring

Video description. Use @image1…@image9 to reference images, @video1…@video3 for videos, and @audio1…@audio3 for audio. Use @character:<request_id> for a Seedance 2 character sheet or @omni-character:<char_id> for a trained Kinovi character. Multiple characters are supported.

Default Value@image1 is the main character. The person walks along a city street at sunset, cinematic lighting.

Image URLsarray

Up to 9 reference image URLs (JPEG/PNG/WebP). Each Nth image corresponds to @imageN in the prompt.

Default Valuehttps://d3adwkbyhxyrtq.cloudfront.net/ai-images/186/712345784292/4a8c5c70-abcc-4920-873e-b0e219986453.jpg

Video Reference URLsarray

Up to 3 reference video clip URLs (MP4, max 15s each). Each Nth video corresponds to @videoN in the prompt.

Default Valueundefined

Audio Reference URLsarray

Up to 3 reference audio files (MP3/WAV, total max 15s). Each Nth audio corresponds to @audioN in the prompt.

Default Valueundefined

Aspect RatioEnum (6 options)

Output video aspect ratio.

Default Value16:9

Duration (seconds)int

Video duration in seconds.

Default Value5

📖

Implementation Guide

Developer documentation

Prepare and upload your reference materials: up to 9 images via images_list, up to 3 videos via video_files, and up to 3 audio files via audio_files.
Craft a detailed prompt that uses clear reference notation: '@image1 as the primary visual style, @video1 as background motion, @audio1 as the lead voiceover, @audio2 as ambient music.' Be specific about how references should be weighted and combined.
Choose your aspect_ratio for the target format (e.g., '16:9' for cinema, '9:16' for vertical, '1:1' for square).
Set your duration between 4 and 15 seconds using the duration parameter.
Submit—the model will generate a 4K masterpiece that synthesizes all your reference materials into a visually stunning, sonically coherent output.

❓

Common Questions

Frequently asked

Does 4K processing time take significantly longer?

4K does require more computational resources, but muapi's infrastructure is optimized for efficient 4K generation. Processing time is reasonable for a production tool—typically completing within a couple of minutes depending on complexity and system load.

What file formats and sizes work best for references?

Standard formats (JPEG/PNG for images, MP4/WebM for video, WAV/MP3 for audio) are supported. For best results, provide high-quality source materials—1080p or higher for images, high-bitrate video, and clear audio tracks. The model will handle the synthesis efficiently.

What is the per-generation cost for Seedance 2 Omni Reference VIP 4K?

This model costs $6.75 per generation. The price is fixed regardless of reference complexity—whether you use 3 materials or all 15 maximum, you pay one transparent per-generation rate.