Chinese AI startup Shengshu Technology, in partnership with Tsinghua University, unveiled the country’s latest text-to-video model Vidu last weekend, which it said could generate videos no longer than 16 seconds with 1080p resolution. Vidu is “the first video model globally to make a major breakthrough” since the release of OpenAI’s Sora in February, the Beijing-based company said on its official WeChat post, while adding that Vidu’s performance “fully matches” that of its international counterparts. Shengshu Technology is the latest AI firm touting its ability to build a Chinese program akin to Sora, which is able to produce 60-second-long videos based on prompts. OpenAI has yet to make the program available to the public. [Shengshu Technology, in Chinese]
Related
