Blog

This post is a follow-up to our first entry in the multi-series blog focused on how…

PyTorch* 2.6 has just been released with a set of exciting new features including torch.compile compatibility…

Meta: Hongtao Yu, Manman Ren, Bert Maher, Shane NayNVIDIA: Gustav Zhu, Shuhao Jiang Over the…

We are excited to announce the release of PyTorch® 2.6 (release notes)! This release features…

2024 has been a year of incredible growth for PyTorch. As that continues in 2025,…

Intel has long been at the forefront of technological innovation, and its recent venture into…

Large Language Models (LLMs) are typically very resource-intensive, requiring significant amounts of memory, compute and…

This blog is the fifth in a series focused on accelerating generative AI models with…

In this blog, we will briefly introduce torchtune, the Ascend backend, and demonstrate how torchtune…

We are excited to announce the addition of embedding operators with low-bit weights (1-8 bit)…

Large Language Models (LLMs) are trained on vast volumes of data and use billions of…

We are pleased to officially announce torchcodec, a library for decoding videos into PyTorch tensors. It…

2D block quantization for Float8 (FP8) holds the promise of improving the accuracy of Float8…

IBM: Krish Agarwal, Rishi Astra, Adnan Hoque, Mudhakar Srivatsa, Raghu GantiMeta: Less Wright, Sijia Chen…

IBM: Tuan Hoang Trong, Alexei Karve, Yan Koyfman, Linsong Chu, Divya Kumari, Shweta Salaria, Robert…

In this blog, we present a case study on distilling a Llama 3.1 8B model…

Figure 1. FP8 GEMM Throughput Comparison CUTLASS vs Triton Summary In this post, we provide…

The vLLM engine is currently one of the top-performing ways to execute large language models…

The Triton open-source programming language and compiler offers a high-level, python-based approach to create efficient…

Introduction At the recent PyTorch Conference, Arm highlighted the widespread impact of its technology, spanning from…