NVIDIA and ARM announced in the GPU Technology Conference (GTC) their alliance to bring deep learning inference to billions of electronic gadgets for retail customers and Internet devices of things to be launched into the global marketplace.
According to the agreement, NVIDIA and ARM will integrate the architecture of the open source Nvidia deep-learning accelerator (NVDLA) into Arm’s Project Trillium platform for machine learning. The collaboration will make it easier for IoT chip manufacturers to integrate artificial intelligence into their designs and to sell affordable and intelligent products to billions of customers worldwide
“Inference will be an essential capacity of each IoT device in the future,” said Deepu carving, vice president and general manager of standalone machines at NVIDIA. “Our partnership with ARM will drive this wave of adoption, as it allows hundreds of chip manufacturers to incorporate deep learning technology.”
“Accelerating artificial intelligence on the boundary is critical to making arm’s vision of connecting a billion IoT devices a reality,” said Rene Haas, executive vice president and chairman of the IP group at arm. “Today, we take another step towards that vision by incorporating NVDLA into the ARM Project Trillium platform, for our entire ecosystem to benefit immediately with the experience and capabilities that our two companies can bring to artificial intelligence and IoT. “