The generative AI industry is trending more than aver, thanks to the rise of Large Language Model, and the war that happens.
After OpenAI popularized this with the introduction of OpenAI, two could play this game. Then three, four, five, and more followed, each developing their own products and solutions.
While most products come from the West, the East isn't falling behind.
China knows well how lucrative this generative AI trend is, and one of the leading tech companies in the country has debuted a text-to-video generative AI to compete directly against the likes of OpenAI Sora, Luma AI's Dream Machine and more.
After the likes of Kling AI and DynamiCrafter, the generative AI industry welcomes yet another product.
MiniMax, the Chinese company based in Shanghai, introduces what it calls the 'Video-01'.
The text-to-video service just goes beyond than just realistic.
今天是 MiniMax 第一次开线下发布会。
阵容强大，视频、声音、文本模型以及产品团队的一线负责人都来现场讲解。
不得不说线下就是松弛，各种技术问题来之不拒，机会属实难得。
整个活动就是一场大 party，和几十位老同事聊天，不知不觉就聊到了很晚。
同甘不共苦，也是一种快乐～ pic.twitter.com/oT6tRMcspr— orange.ai (@oran_ge) August 31, 2024
Founded in December 2021, MiniMax has created AI text and music-generating features, as part of its consumer-facing Hailuo AI platform
This time, the Video-01 text-to-video-generating tool, is a free text-to-video service into the rapidly expanding text-to-video technology market.
The company, known as one of China’s AI “tigers” made Video-01 available to the public via its website after unveiling the tool at the company’s first developer conference in Shanghai.
Video-01, which supports both English and Chinese, can create six-second videos in about a two-minute processing time, and offers various styles, including "cartoon" and "cyberpunk."
In all, the AI model supports 1280x720 resolution videos at 25 frames per second.
Soon after its debut, Video-01 quickly makes waves for its ability to generate hyper-realistic footage of humans that is consistent, including accurate hand movements, which is something most others have struggled with.
In one of the official demo, MiniMax showcased a trailer for a magical adventure where a child touches a coin and is transported through history.
It features special effects and other realism, all of which were made from just text prompts.
MiniMax founder and CEO Yan Junjie said that Video-01 is the first iteration of the firm’s video-generating tool.
He pointed out that MiniMax plans to expand the service to include longer videos and features like picture-to-video conversion, and the ability to edit these videos.
MiniMax 视频生成模型的官方 Demo，讲述一个小男孩拥有魔法硬币的故事。
雨林沙漠，天空大海，魔法科幻，古代未来。
1920*1080分辨率，原生25帧率，全部 AI 生成。
在现场的大屏幕上看完，全场欢呼鼓掌，真的有种大片的冲击感。
尽管现在仍在技术早期，尽管每个场景只有6秒，尽管制作一个2分钟的… pic.twitter.com/kqvtI3URMw— orange.ai (@oran_ge) September 1, 2024
Here, MiniMax becomes one of a handful of Chinese companies that have been making aggressive moves into this nascent AI market segment, which was significantly heated up following OpenAI's revelation of Sora.
After making its tool public, MiniMax's AI technologies handle at least 3 billion daily AI requests across its own and partner platforms worldwide.
And just like other generative AI platforms, MiniMax also sees a significant increase in usage.
The Alibaba and Tencent-backed unicorn startup said that its AI usage time has reached 53% of that of ChatGPT, which is a significant increase from just 3% in 2023, as noted by Yan Junjie.
"We have indeed made significant progress in video model generation, and based on internal evaluations and scores, our performance is better than that of Runway in generating videos," said Yan.