Nvidia has sold many of its computer chips to every major company that makes artificial intelligence models. But now, as public interest in open source and do-it-yourself AI grows, the company has announced that it will also begin offering a “personal AI supercomputer” later this year, starting at $3,000, that anyone can use. own home or office. Nvidia’s new desktop machine, dubbed Digit, will go on sale in May and is about the size of a small book. It contains the Nvidia “superchip” called GB10 Grace Blackwell, optimized for the computational acceleration needed to train and run AI models, and is equipped with 128 gigabytes of integrated memory and up to 4 terabytes of NVMe storage to handle particularly large AI programs. Jensen Huang, founder and CEO of Nvidia, announced the new system, along with several other AI offerings, during a keynote speech today at CES, the annual confab for the computer industry held in Las Vegas. (You can check all the biggest announcements on the WIRED CES live blog.) in a statement released ahead of the keynote.Nvidia said the Digits machine, which stands for “deep learning GPU Intelligence training system,” will be able to run large language models from one to 200 billion parameters, a rough measure of the complexity and size of the model. To do this today, you have to rent space from a cloud provider like AWS or Microsoft, or build a custom system with some chips designed to run AI. If two Digits machines are connected using a proprietary high-speed interconnection link, Nvidia says it will be able to run the best version of Meta’s open source Llama model, which has 405 billion parameters. Digit will make it easier for enthusiasts and researchers. to experiment with models close to the basic capabilities of OpenAI’s GPT-4 or Google Gemini in the office or basement. But the best version of that ownership model, housed in giant data centers owned by Microsoft and Google, is likely bigger and more powerful than anything Digit can handle. Nvidia has been one of the biggest beneficiaries of the AI boom. Its stock price has soared over the past few years as tech companies have been clamoring to buy large quantities of the advanced hardware chips they produce, a key ingredient in developing sophisticated AI. The company has proven adept at creating hardware and software optimized for AI, and its product roadmap has become an important signal of where the industry is expected to lead next. When released, Digits will be the most powerful consumer computing hardware Nvidia offers. It has sold a variety of chipsets for AI development known as Jetson that start at around $250. These can run smaller AI models and can be used like mini desktop computers or installed in robots to test different AI programs.
