At the GPU technology conference in Tokyo, the founder and CEO of Nvidia, Jensen Huang, announced the new Nvidia Tesla T4 GPU and the TensorRT, a software to enable intelligent voice, video, image and Recommendation Services.
The announcement on the new Nvidia TensorRT Hyper-scale platform based on the latest GPU architecture from Nvidia Turing was among the news that Huang shared, addressing the automotive, robotics and healthcare industries, in a talk of two Hours that addressed almost every aspect of the deployment of Artificial intelligence on a global scale.
“There is no doubt that the artificial intelligence of deep learning is being deployed around the world, and we are seeing incredible growth here,” Huang said to an audience of more than 4,000 journalists, partners, academics and technologists gathered in the latest GTC World Tour Stop.
In this way, the CEO of Nvidia, Jensen Huang, introduced the Tesla T4: “A universal Inference accelerator” based on the new Nvidia’s Turing GPU architecture.
The GPU T4 and NVIDIA’s TensorRT software promise to process queries that drive these services faster than any competing platform, up to 40 times faster than CPU alone.
To help companies that support deep learning Web services with the GPU, Huang also introduced the Hyper-scale TensorRT platform that allows a GPU-driven server to process multiple models and deep learning frames at the same Time.
“As a result, utility and utilization increase a lot,” Huang said. “If each node can run any model at the same time, then the utilization of this server will be maximum”
Huang also announced NVIDIA AGX, a series of integrated high-performance Artificial intelligence computers built around the New Xavier processors from NVIDIA, the world’s first processors built for autonomous machines.
Huang announced the availability of several kits for developers that will allow them to quickly put to work Xavier: the Jetson AGX Xavier Devkit for autonomous machines, like robots and drones, and the DRIVE AGX Xavier Devkit for autonomous vehicles.
Huang also announced that Yamaha Motor Co. He selected NVIDIA Jetson AGX Xavier as the development system for his next line of autonomous machines.