Shengshu's Vidu Q4 Preview debuts at 0.09 yuan/sec, betting on character performance and VFX
新智元 · wechat · 2026-10-15
Shengshu Technology launched Vidu Q4 Preview, emphasizing character acting, camera work, and complex VFX at a promo price from 0.09 yuan/sec (API: 0.6 yuan/sec at 720P, 0.75 at 1080P), with up to 15 reference images, 3 audio refs, 16-second clips, and 2K/4K output.
- Already used in viral AI shorts; Shengshu is chief tech partner for theatrical AI film 'Sanxingdui: Future Past'
- Demos show dialogue-free emotional acting, rooftop parkour camera work, and energy VFX — a distinctive 'Vidu style' in anime and action scenes
- Competes with Kling (fine control) and Seedance 2.5 (long narrative) as all video models shift from generating frames to directing content
- Backed by Shengshu's world-model research (Motus, Motubrain — #1 on RoboTwin2.0 and WorldArena) unifying video, language, and action
Cheap trial-and-error is pushing AI video toward its 'DeepSeek moment'.
More from Multimodal
- MiniMax-H3 hailed as favorite open release of the year, runs fully local — AdinaYakup · 2026-10-23
- Michal Malewicz experiments with depth maps in his creative workflow — michalmalewicz · 2026-10-23
- One-Shot Tunable Style Transfer on Flux3 Via Reference Latent + Node — Winter_unmuted · 2026-10-23
- PhysVidBench accepted at EMNLP 2026: do video gen models grasp physical commonsense? — aykuterdemml · 2026-10-23
- EMNLP 2026 paper teaches multimodal LLMs to get cartoons via incongruity-resolution supervision — aykuterdemml · 2026-10-23
- A gorgeous Midjourney style: sref code 820221253 for dreamy lavender watercolor looks — umesh_ai · 2026-10-23