Intel and AMD announced on October 20, 2026, that they are merging their separate x86 CPU optimization efforts for the PyTorch AI framework into a single open-source library, oneDNN. This initiative aims to provide a unified software path for AI workloads on both Intel and AMD processors.
As part of the agreement, AMD will integrate its ZenDNN optimizations into oneDNN and join the UXL Foundation as a steering member. The UXL Foundation oversees the development and direction of the oneDNN library.
The collaboration, facilitated by the x86 Ecosystem Advisory Group formed in October 2024, intends to simplify the development and deployment of AI applications on x86 hardware. It focuses on AI inference, edge computing, and latency-sensitive tasks.
The companies plan to deliver new chip capabilities, such as AI Compute Extensions, through regular updates to oneDNN. This approach is intended to complement existing GPU acceleration rather than replace it, while strengthening the position of x86 architectures against Arm-based competition.
