The new Tensor Float 32 (TF32) precision delivers up to 5x the training throughput of the previous generation and accelerates the training of AI and data science models without code changes. Supporting structural sparsity in hardware doubles inference throughput. Tensor Core also brings AI into graphics processing through features such as DLSS, AI noise reduction, and enhances editing for specific applications.
Learn MoreContact Address:
503 of Ziguang Building, East Gate, Tsinghua University, Haidian District, Beijing
©2022 UNISPLENDOUR XIAOTONG TECHNOLOGY CO., LTD. Privacy Policy
Contact Phone:
Focus on us: