Nvidia's powerful H100 GPU will be available in October

A press document showing the Nvidia H100 Hopper GPU and its applications.Enlarge / A press document showing the Nvidia H100 Hopper GPU and its applications. Nvidia

During today's GTC keynote, Nvidia announced that its H100 Tensor Core GPU is in full production and technology partners including Dell, Lenovo, Cisco, Atos , Fujitsu, GIGABYTE, Hewlett-Packard Enterprise and Supermicro will begin shipping products built around the H100 next month.

The H100, which is part of the "Hopper" architecture, is the most powerful AI-focused GPU ever from Nvidia, surpassing its previous high-end chip, the A100. The H100 includes 80 billion transistors and a special "Transformer Engine" to speed up machine learning tasks. It also supports Nvidia NVLink, which links GPUs together to multiply performance.

According to Nvidia's press release, the H100 would also offer efficiency benefits, delivering the same performance as the A100 with 3.5 times better power efficiency, 3 times lower cost of ownership, and using 5 times fewer server nodes.

Nvidia expects the H100 chip to be used in a variety of industrial, healthcare, supercomputing, and cloud applications, ranging from large language models, to drug discovery, to recommendation, conversational AI, etc. Based on the track record of the older A100 "Ampere" architecture GPU, analysts believe the H100 chip is likely to have a big impact in the AI ​​space. It will also most likely play a role in the next generation of image synthesis models.

Nvidia has announced that more than 50 H100-based server models from different companies will hit the market by the end of the year. And Nvidia itself will begin integrating the H100 into its Nvidia DGX H100 enterprise systems, which contain eight H100 chips and offer 32 petaflops of performance.

News that the H100 is in full production and available soon can be seen as a relief to Nvidia investors. The company's shares fell in early September after it announced that the US government was imposing export restrictions on the H100 to China, where the chip was also partially developed and manufactured. The United States has allowed development of the H100 to resume, but export restrictions to China and Russia are still in place.

Nvidia's powerful H100 GPU will be available in October
A press document showing the Nvidia H100 Hopper GPU and its applications.Enlarge / A press document showing the Nvidia H100 Hopper GPU and its applications. Nvidia

During today's GTC keynote, Nvidia announced that its H100 Tensor Core GPU is in full production and technology partners including Dell, Lenovo, Cisco, Atos , Fujitsu, GIGABYTE, Hewlett-Packard Enterprise and Supermicro will begin shipping products built around the H100 next month.

The H100, which is part of the "Hopper" architecture, is the most powerful AI-focused GPU ever from Nvidia, surpassing its previous high-end chip, the A100. The H100 includes 80 billion transistors and a special "Transformer Engine" to speed up machine learning tasks. It also supports Nvidia NVLink, which links GPUs together to multiply performance.

According to Nvidia's press release, the H100 would also offer efficiency benefits, delivering the same performance as the A100 with 3.5 times better power efficiency, 3 times lower cost of ownership, and using 5 times fewer server nodes.

Nvidia expects the H100 chip to be used in a variety of industrial, healthcare, supercomputing, and cloud applications, ranging from large language models, to drug discovery, to recommendation, conversational AI, etc. Based on the track record of the older A100 "Ampere" architecture GPU, analysts believe the H100 chip is likely to have a big impact in the AI ​​space. It will also most likely play a role in the next generation of image synthesis models.

Nvidia has announced that more than 50 H100-based server models from different companies will hit the market by the end of the year. And Nvidia itself will begin integrating the H100 into its Nvidia DGX H100 enterprise systems, which contain eight H100 chips and offer 32 petaflops of performance.

News that the H100 is in full production and available soon can be seen as a relief to Nvidia investors. The company's shares fell in early September after it announced that the US government was imposing export restrictions on the H100 to China, where the chip was also partially developed and manufactured. The United States has allowed development of the H100 to resume, but export restrictions to China and Russia are still in place.

What's Your Reaction?

like

dislike

love

funny

angry

sad

wow