Name: torch.accelerator: A Unified, Device-Agnostic Runtime API for Stream-Based Accelerators - Yu Guangye, Intel
Start: 2025-06-07T11:10:00+0800
End: 2025-06-07T11:30:00+0800

In-person | 2025 June 7
Learn more on our website

The Sched app allows you to build your schedule but is not a substitute for your event registration. You must be registered to participate in the sessions. If you have not registered but would like to join us, please visit the BAAI Conference webpage.

Please note: This schedule is automatically displayed in China Standard Time (UTC+08:00). To see the schedule in your preferred timezone, please select from the drop-down located at the bottom of the menu to the right.

IMPORTANT NOTE: Timing of sessions and room locations are subject to change.

Saturday June 7, 2025 11:10 - 11:30 CST

TBA

Motivation
PyTorch supports a wide range of acceleration hardware beyond CPUs, including CUDA, XPU, MPS, NPU, HPU, and more. Its architecture allows new backend integration through two key components: ATen operators and device runtime.
While ATen operators are device-agnostic, the runtime remains device-specific, relying on APIs like torch.cuda and torch.xpu. This fragmentation complicates writing portable, hardware-agnostic code across PyTorch and its ecosystem.
To address this challenge, we propose torch.accelerator: a unified, device-agnostic runtime API for stream-based accelerators.
Design
An Accelerator refers to a device that collaborates with the CPU to accelerate computation, typically via asynchronous execution using Stream and Event for synchronization. Our design assumes a single active accelerator per host.
The torch.accelerator API provides a consistent interface for device and stream management, with backend support integrated via the existing DeviceGuardImplInterface registration mechanism.
Further Work
We are actively working on a unified device memory API. These will streamline the library, model, and UTs.
Reference
https://github.com/pytorch/pytorch/pull/132204

Speakers

Yu Guangye

AI Framework Engineer, Intel

An AI framework engineer at Intel, dedicated to supporting Intel GPUs in PyTorch and advancing the generalization and improvement of PyTorch and its ecosystem. With deep experience in SYCL, XPU backend integration, and performance optimization across heterogeneous platforms, I’ve... Read More →

Saturday June 7, 2025 11:10 - 11:30 CST
TBA

Core PyTorch Framework

PyTorch Day China 2025

Yu Guangye

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!