NVIDIA Unveils Its New AI Chip
Los Angeles, CA – Jensen Huang, founder and CEO of NVIDIA, announced on Tuesday at SIGGRAPH, the world’s premier computer graphics conference, its latest AI chip, GH200. The new GH200 chip features the same GPU as the current flagship H100 model but delivers up to 3.5x more memory capacity and 3x more bandwidth.
The new chip aims to bolster its position in the competitive AI hardware market against AMD, Google, and Amazon. Although Nvidia currently holds an estimated 80% market share in AI chips, its dominance is challenged due to GPU shortages driven by high demand from tech giants, cloud providers, and startups seeking to develop their AI models.
“To meet surging demand for generative AI, data centers require accelerated computing platforms with specialized needs,” said Huang. “The new GH200 Grace Hopper Superchip platform delivers this with exceptional memory technology and bandwidth to improve throughput, the ability to connect GPUs to aggregate performance without compromise, and a server design that can be easily deployed across the entire data center.”
“The inference cost of large language models will drop significantly,” according to Huang. This chip advancement will notably benefit models like ChatGPT, a widely used language generation AI.
The chip is set to be available through Nvidia’s distributors in the second quarter of next year.
