PulseAugur
EN
LIVE 06:16:11
中文(ZH) 阿里视频大模型Wan3.0开启公测:文档、ppt也能变视频

Alibaba launches Wan 3.0 video model with expanded input formats

Alibaba has launched a public beta for its video generation large model, Wan 3.0. This upgraded model can now produce videos up to 30 seconds long and supports a wider range of input formats, including documents like PPTs and PDFs, in addition to text, images, and audio. Wan 3.0 aims to enhance realism in video generation, focusing on detailed human features and natural expressions, and also offers improved video editing capabilities. AI

IMPACT Enhances realism and versatility in AI video generation, enabling new applications for document-to-video content.

RANK_REASON Frontier-lab model release with system card [lever_c_demoted from frontier_release: ic=1 ai=1.0]

Read on 雷峰网 (Leiphone) →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Alibaba launches Wan 3.0 video model with expanded input formats

COVERAGE [1]

  1. 雷峰网 (Leiphone) TIER_1 中文(ZH) ·

    Alibaba's large video model Wan3.0 opens public beta: documents and PPTs can also become videos

    <p style="text-align: center;"><img src="https://static.leiphone.com/uploads/new/images/20260807/6a75419d20d2d.jpeg?imageView2/2/w/740" /></p><p>8月6日,阿里巴巴视频生成大模型Wan 3.0开启公测,在生成时长、万能创作、全能参考以及真实世界还原等维度全面升级:单次可生成30秒视频,完整表达创作意图;在文本、图片、音频、视频四种基础模态之外,首次支持doc、xls、ppt、pdf、md等文档格式输入;力求准确还…