The Tech: Its Wafer-Scale Engine (WSE-3) packs 4 trillion transistors and 900,000 AI cores onto a single piece of silicon. [1]
Market Position: Designed primarily for low-latency AI inference, it is up to 15 times faster than standard GPUs for certain large-model workloads. [1, 2]
Catalysts: Backed by a $20 billion exclusive inference capacity contract and a $1 billion loan with OpenAI, as well as an infrastructure deployment deal with Amazon Web Services (AWS). [1, 2]