Model
Seedance 2.0
ByteDance's video generation model released in February 2026, built to produce picture and sound together from text, image, audio or video inputs, and known for a copyright dispute that paused its global rollout.
Status AvailableModality Text, Image, Audio, Video Era Generative AI for everyone
- License
- Proprietary
- Release date
- Developer
- ByteDance
What it is
Seedance 2.0 is a video generation model released by ByteDance on 12 February 2026.[1][2] Rather than treating sound as a separate, later step, it is built on a unified multimodal architecture that generates audio and video jointly, so that picture and sound are produced together instead of being stitched together afterwards.[3]
How it works
The model accepts text, images, audio and video as inputs, and images, audio and video can serve as references that guide a generation.[3] Within a prompt, users can direct performance, lighting, shadows and camera movement — the elements of a shot that a director would normally control on set.[3] ByteDance describes the model as offering the most comprehensive multimodal content reference and editing capabilities in the industry, and as showing exceptional motion stability with ultra-realistic, immersive audio-visual output.[3] Those are the company's own characterisations.
ByteDance evaluated Seedance 2.0 on SeedVideoBench-2.0, its internal benchmark, covering text-to-video, image-to-video and multimodal tasks.[3] The company says the model holds a leading position across dimensions on that benchmark, but its page gives no numeric scores, so the claim cannot be checked from outside.[3]
Release and viral clips
Soon after release, clips made with the model spread widely online. Among them were a fight between figures resembling Brad Pitt and Tom Cruise, and a version of the characters from Friends reimagined as otters.[4] Such clips travelled because the motion looked plausible and the sound matched the picture, and because placing recognisable faces and characters into new scenes had become easy.
The copyright dispute
That ease brought a reaction. The Motion Picture Association denounced the model for copyright infringement; Disney sent a cease-and-desist letter on 13 February 2026, and Paramount Skydance followed around 15 February.[4] On 16 February 2026 ByteDance said it respected intellectual property rights and would strengthen safeguards, and it later paused the model's global launch, according to The Information.[4][1]
Pressure continued. On 16 March 2026, U.S. Senators Marsha Blackburn and Peter Welch wrote to CEO Liang Rubo urging ByteDance to shut Seedance down.[4][1]
Place in the line
Seedance 2.0 was succeeded by Seedance 2.5 in July 2026.[4] Its time as the current model was short, but the question it raised — where a generation model should draw the line between capability and rights — carried forward to what came next.
CapabilitiesMultimodal
TechniquesGenerative models
ApplicationsCreative tools
Ecosystem & regionChina
Governance & standardsSafety & governance
Version history
4-
2025
-
Jun 11, 2025
Seedance 1.0
Seedance 1.0 is a video-generation model from ByteDance Seed, announced in June 2025. One model handles text-to-video and image-to-video, produces 1080p multi-shot clips, and ranked first on two Artificial Analysis leaderboards.
-
Dec 18, 2025
Seedance 1.5 pro
An audio-video generation model from Volcano Engine, announced on 18 December 2025, that produces picture, voices, music and ambient sound together and keeps them in millisecond-level sync.
-
Jun 11, 2025
Seedance 1.0
-
2026
-
Feb 12, 2026
Seedance 2.0 · This page
ByteDance's video generation model released in February 2026, built to produce picture and sound together from text, image, audio or video inputs, and known for a copyright dispute that paused its global rollout.
-
Jul 31, 2026
Seedance 2.5
Seedance 2.5, released by ByteDance in July 2026, is an audio-video joint generation model that makes clips of up to 30 seconds per pass, accepts many multimodal references, and supports editing parts of a scene without regenerating the whole clip.
-
Feb 12, 2026
Seedance 2.0 · This page
20262
-
ReleaseByteDance releases Seedance 2.5
ByteDance released Seedance 2.5 in July 2026. The model generates up to 30 seconds of video per pass, offers a beta long-video mode, and can edit part of a scene without regenerating the whole clip. It is offered through an API.
-
ReleaseByteDance releases Seedance 2.0, prompting a Hollywood copyright backlash
ByteDance's Seedance 2.0 video model went live on 12 February 2026. Viral clips drew copyright complaints from the Motion Picture Association and Hollywood studios, and ByteDance later paused the global launch, according to The Information.
Developed by1
Successor of1
Succeeded by1
Notes and sources
- Senators urge ByteDance to shut down Seedance AI video apphttps://www.cnbc.com/2026/03/17/bytedance-seedance-shut-down-tiktok-marsha-blackburn-peter-welch.html
- ByteDance Unveils Doubao 2.0 AI Model to Tackle Complex Taskshttps://www.caixinglobal.com/2026-02-15/bytedance-unveils-doubao-20-ai-model-to-tackle-complex-tasks-102414865.html
- Seedance 2.0https://seed.bytedance.com/en/seedance2_0
- Seedancehttps://en.wikipedia.org/wiki/Seedance_2.0
Cite this page: Seedance 2.0. TellWorld, https://www.tellworld.net/models/seedance-2-0. Updated 2026-11-08