# GPU [[Pascal Architecture]] [[Ampere Architecture]] A100 > P100 A100 and V100 are both GPUs from NVIDIA. The A100 is the successor to the V100 and is based on NVIDIA’s Ampere architecture, while the V100 is based on the Volta architecture12. According to benchmarks, for training convolutional neural networks with PyTorch, the Tesla A100 is 2.2x faster than the V100 using 32-bit precision and 1.6x faster using mixed precision1. For training language models with PyTorch, the Tesla A100 is 3.4x faster than the V100 using 32-bit precision and 2.6x faster using mixed precision