Seedance 2.0

ByteDance's video model with a Fast/High switch, native audio, up to 7 reference images, and director-level camera control.

Max Duration: 15 seconds max
Resolution: 480p, 720p
Aspect Ratios: 16:9, 9:16, 1:1, 4:3, 3:4, 21:9
Audio: Native audio

Up to 720p resolution15 sec max durationNative audioUp to 7 ref images

Create with Seedance 2.0 View Pricing

What makes Seedance 2.0 unique

Multi-image reference support

up to 7 reference images for consistent characters and scenes

Optional first-frame and last-frame control for precise scene anchoring

Native synchronized audio generation

Director-level camera control and cinematic shot composition

Realistic physics simulation

collisions with weight, fabric tearing, believable action sequences

Native CapCut integration

the same engine behind TikTok's video tools

Best for

Advertising and film productionSocial media marketing via CapCutComplex multi-reference projectsCinematic action contentASMR and audio-rich content

Prompt tips

1.
Use multiple reference images for consistent character identity across shots.
2.
Specify camera movements explicitly — Seedance 2.0 follows director-style instructions well.
3.
Include audio cues in your prompt for synchronized sound design.
4.
For action scenes, describe physics and weight to leverage its realistic simulation engine.

How Seedance 2.0 works

Seedance 2.0 uses a Diffusion Transformer (DiT) architecture that replaces the traditional U-Net backbone with a transformer for better scalability. It features a unified multimodal audio-video joint generation system that processes reference images and audio together to produce cinematic output with synchronized sound.

How to use on Flashloop

Step 01

Select Seedance 2.0

Open Flashloop and choose Seedance 2.0 from the model selector in the video creator.

Step 02

Write your prompt

Describe what you want to see in your video — be as detailed as you like.

Step 03

Generate & download

Hit generate and your video will be ready in seconds. Download or share directly.

Frequently asked questions

Seedance 2.0 prioritizes maximum fidelity. Seedance 2.0 Fast is the same model family with a faster, cheaper generation path — better when iteration speed matters more than peak quality.

Yes. It uses a unified multimodal system that generates synchronized audio and video together, not sequentially.

You can attach up to 7 reference images per generation. First-frame / last-frame control is also supported (mutually exclusive with multi-image references).

Yes. Its camera control and CapCut integration make it well suited for advertising and film production workflows.