Chips, sensors, devices and the machines they power, from open-source silicon to AI computers on your desk.
The physical layer: chips, sensors, boards, devices and the machines they add up to. This page covers processors and accelerators, including the open silicon efforts and India's own fabrication plans, along with the consumer and professional hardware that AI is quietly rebuilding — laptops and phones running models locally, desktop machines sold specifically for AI work, and the sensors feeding all of it. We are interested in what the specifications actually mean for what you can do with the thing, which is usually less obvious than the launch material suggests.
NVIDIA announced a 64 GB version of its small DGX Spark AI computer, starting at $4,999. XDA Developers reports that the 128 GB version now often sells for around $7,000, which makes the 64 GB model the cheaper way in. It uses the same GB10 Grace Blackwell Superchip as the 128 GB model, and several units can be linked to pool their memory. NVIDIA says one 64 GB unit can run models of up to 100 billion parameters and two linked units up to 200 billion. It goes on sale on 23 October 2026 through makers including Acer, ASUS, Dell, Gigabyte, HP and MSI.
References: NVIDIA: DGX Spark · XDA Developers