AMD joined the UXL Foundation as a steering member on October 20, 2026, marking a significant step up from its previous status as a general contributor. This move represents a major collaboration between AMD and Intel, two historical competitors in the CPU market. Both companies have agreed to make oneDNN the default open-source library for x86 acceleration in PyTorch, one of the most widely used deep learning frameworks. The goal is to reduce fragmentation across AI software stacks for x86 CPUs and eliminate the need for developers to maintain separate code paths for different chip vendors. The UXL Foundation, established in September 2023 under the Linux Foundation, hosts oneDNN, a library used by PyTorch, TensorFlow, ONNX Runtime and OpenVINO.
Source: Read the original article
