China's MiniMax Takes on OpenAI's Sora in Converting Text to Videos

China's MiniMax Takes on OpenAI's Sora in Converting Text to Videos

MiniMax

 MiniMax, a Chinese AI startup , recently launched a new text-to-video generation model called “video-01,” heating up the competition with other tech companies in China looking to catch up with the progress made by OpenAI’s Sora model.

What is MiniMax Chinese company?

MiniMax, known as one of China’s “tigers” in AI alongside Zhipu AI, Baichuan, and Moonshot AI, has made the video-01 prototype available to the public on its website after launching the new tool at the company’s first developer conference in Shanghai on Saturday. The video-01 prototype allows the user to enter a text description to generate a video up to six seconds in length. The process from text prompt to video generation takes about two minutes.

Yan Junjie, founder and CEO of MiniMax, said at the event that video-01 is the first version of the company's video generation tool, and noted that future updates will enable users to create videos from photos and edit those clips, according to local media reports.

What is video-01 model?

MiniMax’s video-01 model reflects the way Chinese tech companies are making aggressive moves in this emerging field of artificial intelligence, months after OpenAI first revealed samples of videos generated by its Sora model in February and raised concerns about China’s ability to keep up with the U.S. lead. Sora is still not publicly available.

MiniMax

Founded in December 2021, MiniMax is introducing the new video-to-text generation tool as part of its consumer-facing Hailuo AI platform, which also offers AI text and music generation features. Beijing-based startup Shengshu AI launched its video generation tool Vidu in July, with support for both Chinese and English text prompts.

Zhipu AI, a startup valued at over $1 billion like its three AI “tigers” peers, launched a similar tool to Sora in the same month, which accepts both text and image prompts. ByteDance, the owner of TikTok and Douyin, last month launched its new text-to-video generator app Jimeng for download on the Chinese App Store, after launching it on various local Android app stores on July 31. The desktop version was launched in May.

MiniMax

Jimeng lets users create 80 images or 26 videos for free and charges a monthly subscription of at least 69 yuan ($9.70) to create more videos. Alibaba Group Holding said last month it was working on a video generation tool called Tora, based on the OpenSora core model. Alibaba owns the South China Morning Post.

Read also: Lenovo Introduces a Laptop That Moves and Tracks Your Move with the Help of AI and Your Voice

Source

google-playkhamsatmostaqltradent