字节跳动Seedance 2.5将AI视频生成延长至30秒
阅读原文· the-decoder.com字节跳动在火山引擎FORCE大会上发布五款新AI模型。核心产品Seedance 2.5可生成单次30秒视频片段（无需后期拼接），支持场景变化与节奏切换，并同时处理多达50个额外输入（参考图、音频等），后期编辑可保持原有视觉风格。Seedance 2.0升级原生4K与10bit色深。其他模型包括Doubao 2.1 Pro语言模型、Seedream 5.0 Pro图像模型和Seed-Audio 1.0音频模型。据称Doubao 2.1 Pro价格比Claude Opus 4.6低约80%。Seedance 2.5将于7月初上线。
ByteDance's Seedance 2.5 breaks the 30-second barrier for AI video generation
ByteDance introduced five new AI models at Volcano Engine's FORCE conference. The centerpiece is Seedance 2.5, a video model set to launch in early July.
Volcano Engine is ByteDance's cloud business, which offers the company's AI models as services for enterprises. The new model generates single video clips up to 30 seconds long without any post-stitching, complete with scene changes and tempo shifts. It can process up to 50 additional inputs at once - reference images, audio, and more - useful for film scenes with multiple characters. Users can also edit generated videos after the fact while keeping the visual style and look intact.
Seedance 2.0 gets 4K, and four more models join the lineup
Meanwhile, Seedance 2.0 now supports native 4K with 10-bit color depth. Volcano Engine also announced the Doubao 2.1 Pro language model, the Seedream 5.0 Pro image model, and the Seed-Audio 1.0 audio model. According to the company, Doubao 2.1 Pro costs about 80 percent less than Claude Opus 4.6.