ainewsblitz.com

Breaking

Alibaba's Wan3.0 Model Doubles AI Video Length to 30 Seconds, Adds Document-to-Video Input

  • Media Generation
  • Foundation Models

Alibaba's Tongyi Lab has pushed a new flagship video-generation model, Wan3.0, into public beta, doubling the maximum length of a single continuous AI-generated shot to 30 seconds and, for the first time in the Wan line, letting users turn office documents directly into video. The model appeared on Alibaba Cloud's Model Studio and Qwen Cloud under the listing wan3.0-video in the first days of August 2026, ahead of a fuller public unveiling in Hangzhou on August 10, according to Alibaba's own release notes and subsequent trade coverage.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.

$20
Read this article
$29/month
Unlimited — all 27,556 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year