@PyTorch
NVIDIA's (@nvidia) Anjulie Agrusa, Ryan Spring, and Bruce Zitelli will show how the latest NVFP4 pretraining recipes can accelerate large-scale LLM training while closing the quality gap to BF16 at PyTorch Conference North America 2026. Their October 21 session, “Efficient Pretraining of LLMs in NVFP4,” will cover the recipe and PyTorch tooling that make this possible, including recipe design, kernel choices, and the API surface needed to bring NVFP4 training into native PyTorch workflows. The recipes are being upstreamed into the PyTorch ecosystem through TorchAO and TorchTitan, with dense linear NVFP4 training available in TorchAO today. Register by September 4 to save on your conference pass. 🔗 Register: https://t.co/GIZUAha5wA #PyTorchCon