Technical Blog

The greater performance delivered by current-generation NVIDIA GPU-accelerated instances more than outweighs the per-hour pricing differences of prior-generation GPUs. 9 MIN READ

The NCCL 2.12 release significantly improves all2all communication collective performance, with the PXN feature. 8 MIN READ

Learn about GPU Operator 1.9, which includes support for NVIDIA DGX A100 systems with DGX OS and streamlined installation processes. 3 MIN READ

MT-NLG has 3x the number of parameters compared to the existing largest model of this type and demonstrates unmatched accuracy in a broad set of natural language tasks. 13 MIN READ

Despite unprecedented progress in NLP, many state-of-the-art models are available in English only. NVIDIA has developed tools to enable the development of even the largest language models. This post describes the challenges associated with building and deploying large scale models and the solutions to enable global organizations to achieve NLP leadership in their regions. 16 MIN READ

Learn about some of the major optimizations made to the NVIDIA platform that contributed to the nearly 7x increase in performance since the first MLPerf training benchmark. 31 MIN READ