MiniMax, which is financially supported by Alibaba and Tencent, has rolled out a new AI-powered video generator that creates impressively realistic footage, with a particular strength in depicting human hand movements—a common hurdle for many other tools.
Technical Specifications and Capabilities
The newly introduced MiniMax video-01 model offers video resolutions of 1280×720 pixels at a rate of 25 frames per second and can produce six-second clips. Users can describe complex camera movements, and future updates promise to extend clip durations to 10 seconds. The model is attracting attention on social media, both in China and internationally.
Yan Junjie, MiniMax founder, shared that internal tests indicate the tool outperforms Runway in generating videos. The company is now working on version-02, which will include features such as image-to-video and text-and-image-to-video options, alongside the ability to create longer clips.
🚨 BREAKING NEWS 🚨
A Chinese video AI just dropped, and it's already shaking up the video world.Meet #MiniMax, the AI that's set to revolutionize the game and claim its place in the video AI arena.
Check out this 2-minute official demo, created entirely in Text to Video… pic.twitter.com/qldF7M2akS
— Pierrick Chevallier | IA (@CharaspowerAI) September 1, 2024
User Experience and Comparisons
Tom's Guide conducted various tests using the tool with prompts like a futuristic city amid a lightning storm, a butterfly landing on a water lily, and a spaceship launch. Although the tool performed well and is comparable to Luma Labs' Dream Machine, it didn't surpass Runway Gen-3, despite Yan Junjie's claims. However, MiniMax stands out in generating human movement and promises improvements in future versions.
The MiniMax video-01 model is accessible for free, allowing users to explore its functionalities. A demo on the platform X demonstrates a trailer for a magical adventure created using text prompts, AI, and skillful editing. Interested users can try Video-01 free at hailuoai.com/video after registering with a mobile number, including non-Chinese numbers. The company also offers an API for developers.
Company Background and Market Position
Founded in late 2021, MiniMax has already made notable progress in the AI industry. In addition to their new video generator, the company provides a large language model and a text-to-speech model. A recent funding round led by Alibaba accumulated around $600 million, with additional investment from Tencent, valuing MiniMax at approximately $2.5 billion.
MiniMax stresses responsible use of their technology. Users are required to agree to terms of use that prohibit the generation of illegal content, spreading false information, violating privacy, or sharing illegal material. The tool blocks explicit content and includes a subtle watermark on the generated videos.
Market Competition and Future Prospects
While OpenAI's Sora maintains a strong reputation in video AI, several Chinese companies, including KLING, Vidu, and Jimeng AI, have also entered the market. MiniMax's continuous updates and improvements position it as a significant player in the competitive field of AI video generation.
Launched in February, Sora distinguishes itself through its ability to generate hyper-realistic videos from text-based prompts, a process powered by advancements in text-conditional diffusion models. These models are meticulously trained on a library of videos and images, spanning various durations to craft content that simulates real-world motion.
