PyTorch 2.9 expands hardware support with AMD ROCm, Intel XPU, and Arm improvements, offering better multi-GPU programming and performance optimizations for AI developers.
PyTorch 2.9 enhances hardware support for AMD ROCm, Intel XPU, and Arm, improving AI model deployment across diverse ecosystems.
PyTorch 2.9 introduces wheel support for AMD ROCm, Intel XPU, and NVIDIA CUDA 13, with stable ABI for extensions and Symmetric Memory for multi-GPU programming. Developers benefit from AI APIs and SDKs integration.
FlexAttention supports Intel GPUs, flash decoding optimizes CPU backends, and torch.compile offers better error handling. Useful for performance profiling workflows.
PyTorch 2.9 democratizes hardware for ML with AMD ROCm, Intel XPU, and Arm support, enhancing AI model hosting and deployment flexibility.
PyTorch 2.9 adds comprehensive support for AMD ROCm, Intel XPU, and NVIDIA CUDA 13, plus enhanced Arm processor optimizations for broader hardware compatibility.
Symmetric Memory simplifies multi-GPU kernel development by enabling efficient programming across NVLinks and remote direct memory access networks for better performance.
The update brings FlexAttention support on Intel GPUs, flash decoding optimization for CPUs, and enhanced error handling in torch.compile for better development workflows.
The stable libtorch ABI ensures better compatibility for third-party C++ and CUDA extensions, making it easier to integrate and maintain custom extensions.
Flash decoding with FlexAttention enhances parallelism for key-value sequence processing on CPU backends, improving efficiency for certain models.