Current status on July 20, 2026: Dreamina Seedance 2.0 and Seedance 2.0 mini are officially released through BytePlus. The official product page also states that the new API is fully available to clients.
This supersedes the February version of this article, which correctly reflected the documentation at that time but is now outdated.
Looking for Seedance 2.5? Dreamina now says it is live globally, and BytePlus has published the international model entry, API tutorial, model ID, and token pricing. Read the Seedance 2.5 release date, specs, access, and API status.
Seedance 2.0 Release Status
Seedance 2.0 is no longer a rumored version name. BytePlus maintains an official product page, a ModelArk tutorial, a prompt guide, pricing documentation, and video generation API references for the current series.
Verify the current status at these first-party sources:
- BytePlus Seedance 2.0 product page
- Dreamina Seedance 2.0 series tutorial
- ModelArk video generation API
Official Seedance 2.0 Specs
BytePlus describes Seedance 2.0 as a multimodal video model that can use text, images, video, and audio as references. The product page highlights the following capabilities:
- 4-to-15-second generation options;
- 480p, 720p, 1080p, and 4K output options;
- synchronized dialogue, ambient sound, and effects;
- stronger camera-language and framing adherence;
- improved physical motion and object interaction;
- reference consistency across structure, materials, people, and scenes;
- multiple aspect ratios and fixed or dynamic cameras.
Availability can differ by model, plan, account, and region. A capability shown on the family product page may not be valid for every API parameter combination, so confirm the exact request schema before integration.
Seedance 2.0 and Seedance 2.0 Mini
The official release includes two variants. Instead of assuming that one is always better, benchmark both against your delivery requirements.
Seedance 2.0 mini can be a practical iteration model when throughput and experimentation matter. The full model should be tested on hero shots, complex reference combinations, and scenes where motion or structural fidelity determines whether the clip is usable.
For a fair comparison, keep the following fixed:
- prompt and negative constraints;
- input references;
- duration and aspect ratio;
- resolution;
- acceptance criteria.
What Changed from Seedance 1.x?
The meaningful shift is not just the version number. The 2.0 product positioning centers on multimodal reference control, audio-visual alignment, complex physical movement, and narrative camera execution. Those capabilities make it more relevant to multi-shot storytelling than a workflow built around isolated text-to-video clips.
For motion comics and animated stories, the model still does not replace production structure. Character sheets, scene references, shot intent, dialogue timing, and review states should live outside the generation request. That is what lets a team regenerate one shot without losing the rest of the episode.
API and Pricing Availability
ModelArk provides task-based video generation APIs. Pricing is dynamic and depends on the model and generation settings; the live pricing page or console quote should be used for purchasing decisions.
Read the Seedance 2.0 API pricing guide for the integration flow and cost model.
FAQ
Is Seedance 2.0 real and official?
Yes. BytePlus publicly lists Dreamina Seedance 2.0 and 2.0 mini as an official release.
Does Seedance 2.0 have an API?
Yes. BytePlus ModelArk exposes video generation task operations and a dedicated Seedance 2.0 series tutorial.
Can Seedance 2.0 generate audio?
The official product page describes synchronized audio and video, including ambient effects and human speech. Confirm model- and mode-specific support in the live API documentation.
Does Seedance 2.0 support 4K?
The product page advertises 4K among the resolution options. Check plan and request constraints before assuming that it is available for every generation mode.