Alibaba's Wan3.0 turns your slide deck into a 30-second film
Wan3.0 hit public beta with 30-second single-pass clips, one unified model, and PDFs or PPTs as direct video input.
Alibaba pushed Wan3.0 into public beta on 6 August 2026, and the headline number is duration: 30 seconds in a single pass, double the 15-second ceiling of Wan2.7-Video and well past the few-seconds-to-15 range most mainstream generators cap at. Single-pass matters more than the number — it means continuous camera moves and an actual story beat instead of stitched fragments. A smart duration feature picks the optimal length from the prompt.
The input side is the sneakier upgrade. Wan3.0 takes text, images, audio, video, web pages, and documents — doc, xls, ppt, pdf, md — up to 100MB or 50 pages per file. Feed it a pitch deck or a research PDF and it renders video, holding faces, voices, and layouts consistent throughout. Alibaba also collapsed the four separate models Wan2.7 needed into one, and output caps at 1080p.
Access is by application via Alibaba Cloud Model Studio and Qwen Cloud. API pricing starts around $0.05 per second at 480P, with 720P and 1080P tiers above it — cheap enough that document-to-video stops being a demo and starts being a workflow.
Why it matters: doc-to-video at five cents a second turns every deck, report, and changelog into a publishable clip — and puts the squeeze on the entire explainer-video industry.
Sources
- Alibaba Unveils Wan3.0 with Twice as Long Video Outputs from a Richer Variety of Inputs alizila.com
- China's Alibaba releases Wan3.0 AI video model in public beta with 30s clips, multimodal inputs technode.global
Written by an AI pipeline from the sources above. How it works.
Feed, daily deep-dive and bytes — readable offline, with push alerts for the topics you follow.