Guides / Library / Seedance 2.0
Seedance 2.0 on Higgsfield
Video model from Bytedance. Higgsfield's catalog lists it as: Reference-driven video with image/video/audio reference inputs, consistent identity, multi-SKU; optional generate_audio for native audio.
Specs
| Output | Video |
| Made by | Bytedance |
| Accepts | text prompt, start frame, end frame, reference images, reference video, reference audio |
| Clip length | 4 to 15 seconds |
| Resolution | 480p, 720p, 1080p, 4k |
| Modes | std, fast |
| Genre presets | auto, action, horror, comedy, noir, drama, epic |
| Sound | can generate audio with the clip |
| Aspect ratios | auto, 16:9, 9:16, 4:3, 3:4, 1:1, 21:9 |
Similar video models
Seedance 2.0 Mini
Fast budget Seedance 2.0 variant with image/video/audio reference inputs and native audio
MiniMax H3
Multimodal video generation with keyframes or image/video/audio references
MiniMax H3 Max
Fast text-to-video, keyframe, and multimodal-reference video generation
Seedance 2.5
Seedance 2.5 text-to-video, multimodal omni-reference generation, video edit, and video extension
Guides that use this kind of model

Product ads from one photo
Turn a clean product photo into vertical ad clips with Marketing Studio and Ad Multiplier.

One character across many shots
Use Soul ID and Soul Cast so the same face and outfit survive from clip to clip.

Your first Higgsfield video
Go from an empty account to one finished 5 to 10 second clip without burning the month's credits.
Sources
Last checked 18 November 2026. Higgsfield changes models, prices and menus often, so confirm on the live site.
Up to 50% off a Higgsfield plan
Higgsfield gives people who arrive through an affiliate link up to 50% off. The offer runs for three hours from the moment you log in, so pick your plan before you click.
Affiliate link. I earn a commission if you subscribe, at no extra cost to you. Terms checked 18 November 2026.