Back to Gap Map Model components / Image, video, 3D & music generation

Seedance

closed / Overall score: n/a

Seedance is ByteDance Seed's video generation model line, which generates video and synchronized audio in one pass from text, image, audio and video references. It is served in ByteDance's Jimeng and Doubao apps and through the BytePlus ModelArk API, with multi-shot storytelling and editing of existing clips.

Openness

1 high confidence

1.0

weights: closed(no weights are distributed
data: closed
code: closed
license: Proprietary(ByteDance and BytePlus service terms

Seedance runs only in ByteDance's apps and through its cloud API. No weights, training data or code are published.

https://seed.bytedance.com/en/blog/one-take-creation-flexible-referencing-introducing-seedance-2-5 recorded 2026-09-26
"Seedance 2.5 is rolling out on Jimeng AI, Doubao Pro, and other platforms, with API access coming soon via BytePlus ModelArk."
https://seed.bytedance.com/en/seedance2_0 recorded 2026-09-26
Seedance 2.0 page: "a unified multimodal audio-video joint generation architecture", with "Get API" and "Try in Playground" links and no download.

Adoption

not assessed

ByteDance publishes no usage figure for Seedance, and a hosted model has no download count, so no reading was possible.

https://seed.bytedance.com/en/seedance2_0 recorded 2026-09-26
Product page, read for a usage disclosure: none published.

Capability

3 medium confidence

3.0

Seedance 2.0 places near the top of the video arena, though behind the downloadable MiniMax H3. Its newer 2.5 release had not yet been ranked there.

https://artificialanalysis.ai/video/leaderboard/text-to-video recorded 2026-09-26
Without-audio leaderboard entry "Dreamina Seedance 2.0 720p" (rank 6, Elo 1258.8).

Verified 2026-09-26