Nvidia
Latest about Nvidia
Datacenter GPU service life can be surprisingly short
By Anton Shilov published
But while datacenter GPUs can live longer with a lower utilization rate, this hardly makes a lot of economic sense.
Nvidia's Jensen Huang admits AI chip design flaw was '100% Nvidia's fault' — now-fixed Blackwell chips in production
By Anton Shilov published
Nvidia CEO Jensen Huang admitted Wednesday that Nvidia was 100% at fault for a design flaw in the company's new Blackwell AI chips. The flaw has been fixed and the chips are now in production.
Former Nvidia engineer discovers 41-million-digit long prime number
By Jowi Morales published
Former Nvidia engineer uses GPU cloud infrastructure to discover the largest prime number known to man yet.
Microsoft Azure CTO claims distribution of AI training is needed as AI datacenters approach power grid limits
By Anton Shilov published
Microsoft, other AI giants could take a page from Chinese researchers, who trained an AI model using multiple datacenters at once.
Microsoft Azure flaunts world's first custom Nvidia Blackwell racks
By Anton Shilov published
Non-Nvidia servers running Blackwell are up and running.
Nvidia and MediaTek collaborate on 3nm AI PC CPU
By Dallin Grimm published
Years of rumors may finally prove true.
Save up to $40 on Nvidia RTX 4060 and 4070 class GPUs — AMD RX 7000-series also discounted
By Jarred Walton published
We've checked pricing on current generation graphics cards for October Prime Day 2024 to find the best deals. Mainstream GPUs look to be in the sweet spot for savings.
Supermicro's datacenter-scale liquid cooling solution sets the stage for Nvidia Blackwell
By Anton Shilov published
Supermicro develops datacenter-scale liquid cooling solution for next-generation AI servers based on Nvidia Blackwell GPUs.
Jensen Huang is now worth billions more than Intel
By Jowi Morales published
Nvidia CEO Jensen Huang's net worth hits over $109 billion, making him the 11th richest person in the world.
CUDA-beating ZLUDA breathes new life with financial backing from unknown party — pivots to AI workloads across multiple GPU vendors
By Anton Shilov published
ZLUDA, a CUDA translation layer, gets new funding and therefore refocuses to AI software and multiple GPU vendors support.
Stay On the Cutting Edge: Get the Tom's Hardware Newsletter
Get Tom's Hardware's best news and in-depth reviews, straight to your inbox.