Chinese AI firm MiniMax has released a new video-generation model, called H3, that can process text, images, video, and audio. The model can generate videos of up to 15 seconds in 2K resolution with native stereo sound.
The H3 model is aimed at commercial applications, including advertising, e-commerce, product design, and games. MiniMax said generating 2K video would cost less than one-third of mainstream rival products.
MiniMax plans to release H3’s model weights within days, allowing users to download and customize the underlying system. This move extends the open-weight approach adopted by Chinese AI developers into video generation.
Original reporting: Appleton, WI News Feed (HLL/CB) — read the source article.
