TL;DR What this means for you Contents Getting started FP8 comes to MPS Faster language-model decoding Linear layers Mixture-of-experts layers Linear algebra gets a boost Complex-valued computations Small matrices and larger solves Convolutions, padding, and training Softmax and reductions A much smaller memory footprint for video decoding More control with MPS streams More operations work on MPS A sneak peek at PyTorch 2.16 Faster LLM decoding Attention runs entirely on Metal More MPSGraph to Metal migrations A smaller install Wrapping up What's next for us If you run models on your Mac or use MPS as your primary acceleration tool for computing, there's plenty to look forward to in PyTorch 2.15! FP8 arrives on MPS, language-model decoding gets faster paths through linear layers and softmax and new Metal kernels tackle some surprisingly expensive parts of audio, video and training workloads.