Chinese AI startup MiniMax has joined the country’s heated video generation model competition with the launch of its text-to-video model,cat3movie Video-1, which is available to the public through its official website. The tool is capable of producing a video of up to six seconds after inputting a simple text prompt. At least a dozen such models have emerged in China since US startup OpenAI’s Sora model amazed the world in February, illustrating that local artificial intelligence companies are rushing to grab a share of the emerging field. Video-1 currently can only use text to create new videos, according to MiniMax founder Yan Junjie, with functions including image-to-video expected in future iterations. [Yicai, in Chinese]
