Applied Materials and Micron team up to reinvent AI memory
The ChatGPT you use runs on servers powered by memory, DRAM, HBM, NAND. If these two companies figure out how to make that memory more energy-efficient, it directly affects the cost and speed of the models you use every day. The question is how long before this moves out of the labs.
practitioners › The work is happening at Applied Materials' EPIC Center and in Micron's fabs, focused on process and hardware architecture, not on any announced shippable product.
Applied Materials →