"We're delighted by Google's success - they've made great advances in AI and we continue to supply to Google,"
"Nvidia is a generation ahead of the industry - it's the only platform that runs every AI model and does it everywhere computing is done. Nvidia offers greater performance, versatility, and fungibility than ASICs, which are designed for specific AI frameworks or functions."
"We are experiencing accelerating demand for both our custom TPUs and Nvidia GPUs; we are committed to supporting both, as we have for years,"
Meta may deploy Google's TPUs in its datacenters beginning in 2027, prompting a dip in Nvidia's share price. Nvidia publicly praised Google's TPU advances while asserting that Nvidia GPUs are a generation ahead and offer broader performance, versatility, and fungibility than ASICs. Google's seventh-generation TPUs, codenamed Ironwood, can scale to pods containing 256–9,216 chips and compete with Nvidia Blackwell; Nvidia's upcoming Vera Rubin accelerators are faster but lack Google's scale. Google reports accelerating demand for both its custom TPUs and Nvidia GPUs and commits to supporting both. Wider TPU adoption faces obstacles because Google historically leases TPUs through Google Cloud rather than selling chips.
Read at Theregister
Unable to calculate read time
Collection
[
|
...
]