China Unveils 'Vidu' at 2024 Zhongguancun Forum, Rivaling OpenAI's Sora with 16s 1080p Clips

Apr 27, 2024, 02:39 PM

China has introduced a new artificial intelligence video generation model named 'Vidu', developed by Shengshu Technology and Tsinghua University. Unveiled at the 2024 Zhongguancun Forum in Beijing, Vidu is capable of producing 16-second video clips at 1080p resolution with a single click, and can also generate still images or 3D models. The team behind Vidu includes former employees from Tencent, Alibaba, and Byte, and the model's architecture is named 'U-ViT'. This release marks China's entry into the competitive AI video generation market, positioning Vidu as a direct competitor to OpenAI's Sora. The new model has sparked discussions about its capabilities and the future of AI-driven media production.