FULL STORY

Alibaba's Wan 3.0: 30-Second Audio-Video Generation

Alibaba's Wan 3.0 debuted on Magnific with 30-second audio-video generation. Hands-on tests showed multi-shot output from a single prompt, and demos confirmed strong face consistency across long videos.

2026-08-24 ~ 2026-08-25 · 3 episodes · 7 posts

Episode 1 · Alibaba's Wan 3.0 lands on Magnific with 30-second sound-synced video (2026-08-24, 3 posts)

Alibaba's Wan 3.0 video model is now on Magnific, generating 30-second sound-synced videos in a single call with unified text- and image-to-video. Tests show impressive character consistency and stable lip-sync across language switches.

Episode 2 · Wanxiang 3.0 Generates Seven Coherent Camera Shots from One Prompt (2026-08-24, 2 posts)

Tests show Wanxiang 3.0 can generate seven distinct camera moves in one shot from a single prompt, with stable 3D and cartoon styles at reasonable credit costs.

Episode 3 · Magnific Integrates Wan 3.0 for Consistent Long-Video Generation (2026-08-25, 2 posts)

Magnific has integrated the Wan 3.0 model, demonstrating highly consistent facial generation across 22 years, three age stages, and 13 scenes in a single long video, marking a major advance in character consistency for video generation.