| Mobile Web

Search results for Grace Blackwell

AI & Enterprise

Nvidia unveils DGX Spark 64GB to run 100 billion-parameter models locally

Nvidia on Monday unveiled a 64GB configuration of its personal AI supercomputer, Nvidia DGX Spark. The company said the 64GB system can run local agents on the device, enabling on-device processing without relying on the cloud. As workloads grow, two systems can be connected into a single cluster through Nvidia Sync Cluster Assistant without additional setup. It is scheduled to launch in October through major manufacturing partners.

AI & Enterprise

Jensen Huang congratulates OpenAI\'s Astra, says AGI has arrived, next 400,000 GPUs

Nvidia CEO Jensen Huang (젠슨 황) congratulated OpenAI on the launch of its new AI model Astra, saying artificial general intelligence has arrived. He posted on X that Astra was trained on Nvidia chips and wrote that the next step would be 400,000 GPUs coming online. OpenAI unveiled Astra on Sept. 4 and began providing it to some customers. OpenAI executives differed over how to use the term AGI.

AI & Enterprise

Nvidia begins mass production of Groq 3 LPX for AI agents

Nvidia has shifted its Groq 3 LPX inference accelerator for AI agents into mass production. The chip extends Nvidia\'s Vera Rubin data center platform and targets high-speed token generation to reduce decode latency in agent workflows. In an Artificial Analysis benchmark, it produced 3,400 tokens per second running the open-source Gemma 4 31B model with a 100,000-token context window. Nebius Group is among early adopters, and SpaceX was named a new flagship customer.