ByteDance · Video models
Seedance 2.5, and why it isn't in a coding tool
ByteDance's video model generates 30 seconds in a single pass at up to 4K, takes as many as 50 multimodal reference inputs, and can edit one region of a frame without regenerating the shot.
Not wired up
Nothing to sell you here. We do developer tooling; this is a video model and it doesn't connect to a coding agent in any useful way. The page exists because it shipped the same day as MiniMax H3 and people keep asking which one to use.
The numbers
- Built by
- ByteDance
- Released
- Jul 31, 2026
- In HaryAI
- Not wired up
What it takes to use it
It stays closed — no weights. Access is through ByteDance's own surfaces: Jimeng in China, Dreamina inside CapCut elsewhere. If you want it in a pipeline rather than a UI, that's the constraint to plan around — you're building on a hosted product, not a model you can run.
Want this one?
Leave an email and we'll tell you if we wire it up. That's the whole message — there's no newsletter attached to it.
Figures checked 2026-08-06 against vendor pricing pages and launch coverage. Models re-price often — if something here is stale, tell us.