What Is Seedance 2.5? Features, Release Status & How to Access

Seedance is ByteDance's family of AI video generation models — the lineage behind Seedance 1.0, the audio-capable 1.5 Pro, and the widely used Seedance 2.0. The newest step is Seedance 2.5, and unlike the early speculation, we now have concrete details from ByteDance's own announcement.

Is Seedance 2.5 out yet?

Seedance 2.5 was officially unveiled on June 23, 2026 at the Volcano Engine FORCE conference. It is not yet generally available: it launched into a closed (enterprise) beta, and as of mid-July 2026 it's still there. Third-party reports had pointed to the API opening more broadly around July 16 — that window has passed with no public API, model listing, or pricing appearing. The strongest signal of an imminent launch so far is CapCut's official teaser: "Coming soon: Dreamina Seedance 2.5 is arriving on CapCut." ByteDance still hasn't confirmed an exact public date. For the full picture, see the Seedance 2.5 release date.

What Seedance 2.5 adds

From ByteDance's announcement, the headline upgrades over 2.0 are three — and each changes what's actually practical to make:

• Native 30-second clips in a single generation — double 2.0's ~15 seconds, with no stitching. A longer native clip means a full scene that builds, holds, and resolves, instead of a short loop with visible seams where separate generations were joined.
• Up to 50 reference inputs in one prompt — a big jump from 2.0's limit of roughly 9 images + 3 videos + 3 audio clips. More references means tighter control over characters, style, and continuity across a shot.
• Localized, controllable editing — change part of a shot (an object, a background element) without regenerating the whole clip, keeping the original style intact. That's the difference between "re-roll and hope" and actually art-directing a result.

One correction worth making, because many pages get it wrong: native 4K was a Seedance 2.0 update, not a 2.5 feature. ByteDance didn't headline 4K for 2.5, so don't assume "30 seconds and 4K" until it's confirmed.

How you'll access it

Seedance 2.5 is expected to arrive through the same doors as earlier versions, in stages:

• Volcano Engine (Ark) API — direct, programmatic access aimed at businesses and pipelines. This is the enterprise route, and where broader access is expected to open first.
• Dreamina / Jimeng — ByteDance's consumer creation platforms and the most full-featured way for individual creators. Dreamina serves international users; Jimeng (the China app) typically needs a Chinese phone number.
• BytePlus ModelArk — the international API surface.
• Third-party tools — added once the public Volcano Engine API is live.

For a step-by-step once it opens, see how to access Seedance 2.5.

What's still unconfirmed

Two things ByteDance has not locked down: an exact public release date and pricing. Early pricing chatter is extrapolated from Seedance 2.0 (roughly 1 RMB per second of video) — useful as a rough order of magnitude, but not official. Some coverage has also raised content-ownership and copyright questions around 30-second AI clips; expect the usual platform content policies to apply. Treat any specific launch day or price you see elsewhere as unverified until it's official — we update this page as details land.

How to get ready today

The people who produce great work on day one of a new model already have prompts and workflows ready — and prompt skills carry straight over between Seedance versions. So the smart move is to practice on the shipped version now:

1. Build a library of prompts that already work — browse Seedance 2.5 prompts and copy a few to start.
2. Learn the structure Seedance responds to — see how to use Seedance.
3. Practice on Seedance 2.0, available today. Run a prompt on ponpon and keep the outputs you like.

When Seedance 2.5 opens up, you'll already know what works. For how it compares feature-by-feature, see Seedance 2.5 vs 2.0.