NVIDIA has recently announced that its latest DGX Spark personal AI supercomputer will officially go on sale on October 15. This compact device can easily fit on a desk, yet it offers powerful computing capabilities, making it suitable for users to conduct complex AI model training and research.
The DGX Spark is priced at $3,999, which is higher than the initial release price of $3,000. Users can purchase this supercomputer through NVIDIA's official website as well as some partners and retail stores in the United States. The supercomputer is designed to provide data scientists, AI researchers, and students with convenient AI computing power, helping them explore and innovate in the field of AI. Jensen Huang, CEO of NVIDIA, stated during the launch that he hopes to bring AI supercomputers to everyone's desktop, enabling more people to participate in the development of AI.
In addition, multiple custom versions from third-party manufacturers will also emerge in the market, such as Acer's Veriton GN100, which is also priced at $3,999. In addition to powerful performance, the DGX Spark is equipped with NVIDIA's GB10Grace Blackwell super chip, featuring 128GB of unified memory and up to 4TB of NVMe SSD storage. Its computing power can reach ten trillion operations per second, capable of handling AI models with up to 200 billion parameters. Despite this, the Spark maintains an extremely small size, suitable for use with a standard power outlet, hence it is called "the world's smallest AI supercomputer."
Notably, the DGX Spark also has a larger version —— Station, although its release date has not been disclosed yet. It is expected that this device will bring unprecedented AI computing experiences for individual users and research institutions, marking the popularization of AI technology to a broader user base.
Key Points:
🌟 NVIDIA will launch the DGX Spark personal AI supercomputer on October 15, priced at $3,999.
💻 This device is compact and suitable for desktop use, but it has powerful performance, capable of handling complex AI models.
🤖 In addition to NVIDIA's own products, multiple manufacturers have also launched customized versions, further enriching market choices.
