Developers working in artificial intelligence (AI) can access expanded support for Intel GPUs in PyTorch* 2.5, which was recently released with contributions from Intel. GPUs supported include Intel® Arc™ discrete graphics, Intel® Core™ Ultra processors with built-in Intel® Arc™ graphics and Intel® Data Center GPU Max Series¹.

These new features help promote accelerated machine learning workflows within the PyTorch ecosystem, provide a consistent developer experience and support. Application developers and researchers seeking to fine-tune, inference and experiment with PyTorch models on Intel Core Ultra AI PCs will now be able to directly install PyTorch with preview and nightly binary releases for Windows, Linux and Windows Subsystem for Linux 2.

New features include:

• Expanded PyTorch hardware backend support matrix to include both Intel Data Center and Client GPUs.
• The implementation of SYCL* kernels to enhance coverage and execution of Aten operators on Intel GPUs to boost performance in PyTorch eager mode.
• Enhanced Intel GPU backend of torch.compile to improve inference and training performance for a wide range of deep learning workloads.

In addition, PyTorch 2.5 incorporates improvements and new features for the latest Intel data center CPUs. The FP16 datatype is enabled and optimized through Intel® Advanced Matrix Extensions for both eager mode and TorchInductor to enhance inference capabilities on the latest Intel data center CPU platforms, such as Intel® Xeon® 6 processors. The TorchInductor C++ backend is also available on Windows to bring a better user experience to AI developers in Windows environments.

_{¹Intel Data Center GPU Max Series is available only on Intel® Tiber™ AI Cloud}

_{Performance results are based on testing as of dates shown in configurations and may not reflect all publicly available updates. See backup for configuration details.
Intel does not control or audit third-party data. You should consult other sources to evaluate accuracy.}