Nvidia already sells boatloads of computer chips to every major company that builds its own artificial intelligence models. But now, at a time when public interest in open source and DIY AI is skyrocketing, the company has announced that later this year it will also offer a “personal AI supercomputer,” starting at $3,000, that anyone can use in his or her life. own home or office.
Nvidia's new desktop machine, called Digits, will go on sale in May and is about the size of a small book. It contains an Nvidia 'superchip' called GB10 Grace Blackwell, optimized to speed up the calculations needed to train and run AI models, and is equipped with 128 gigabytes of unified memory and up to 4 terabytes of NVMe storage for processing particularly large AI programs.
Nvidia founder and CEO Jensen Huang announced the new system, along with several other AI offerings, during a keynote speech today at CES, an annual computer industry conference in Las Vegas. (You can watch all the biggest announcements on the WIRED CES live blog.)
“Putting an AI supercomputer on the desks of every data scientist, AI researcher and student will enable them to engage and shape the era of AI,” Huang said in a statement ahead of his keynote.
Nvidia says the Digits machine, which stands for “deep learning GPU intelligence training system,” can run a single large language model with up to 200 billion parameters, a rough measure of a model's complexity and size. To do this today, you would have to rent space from a cloud provider like AWS or Microsoft, or build a custom system with a handful of chips designed to run AI. If two Digits machines are connected via a proprietary high-speed connection, Nvidia says they will be able to use the most capable version of Meta's open source Llama model, which has 405 billion parameters.
Figures will make it easier for hobbyists and researchers to experiment in their offices or basements with models that approximate the basic capabilities of OpenAI's GPT-4 or Google's Gemini. But the best versions of these proprietary models, housed in massive Microsoft and Google data centers, are most likely both bigger and more powerful than anything Digits can handle.
Nvidia is one of the biggest beneficiaries of the AI boom. Its stock price has skyrocketed in recent years as tech companies clamored for large quantities of the advanced hardware chips it produces, a crucial ingredient for developing advanced AI. The company has proven itself adept at optimizing hardware and software for AI, and its product roadmap has become a key signal of where the industry is expected to go.
When it comes out, Digits will be the most powerful consumer computing hardware Nvidia offers. It already sells a range of chipsets for AI development, known as Jetson, which start at around $250. These can run smaller AI models and be used as a mini desktop computer or installed on a robot to test various AI programs .
