NVIDIA announces that its generative AI engine DGX GH200 has been put into mass production, supporting the training of large AI models with trillions of parameters

On May 28th, Eastern Time, NVIDIA founder and CEO Jensen Huang announced in his speech at NVIDIA Computex 2023 that the generative AI engine NVIDIA DGX GH200 is now in mass production.

According to NVIDIA's official website, NVIDIA DGX GH200 is a new AI supercomputer that fully connects 256 NVIDIA Grace Hopper superchips into a single GPU, supporting the training of trillion-parameter AI models. It can handle large-scale recommendation systems, generative AI, and graph analysis, and provide linear scalability for giant AI models.

"There is no need to store data in many modules. DGX GH200 makes it easier to train large language models and deep learning recommendation systems." Huang Renxun said. (China Securities Network)