Reported by 3 outlets
The story so far
AI summary of the reporting below · Updated · How we work
AMD and Intel are consolidating their x86 CPU artificial intelligence optimizations for PyTorch into oneDNN, making it the default open-source library across both chipmakers. AMD will merge its proprietary ZenDNN library into oneDNN and join its governing body, the Unified Acceleration Foundation, as a steering member with a $20,000 annual contribution. [1][2]
The change eliminates the need for developers to maintain separate software plug-ins to get tuned performance out of Intel and AMD hardware. Instead, oneDNN detects processor features at runtime to execute optimized kernels. The companies have not disclosed which specific PyTorch release will debut the unified path, nor have they published release timelines or performance figures. [1][2]
The consolidation stems from a recommendation by the x86 Ecosystem Advisory Group, which the rival chipmakers formed in October 2024 alongside tech partners including Microsoft and Google. Intel says the single library approach also lays the software groundwork for AI Compute Extensions, a shared matrix-multiplication instruction set the two chipmakers are co-designing for future processors. [1][2][3]
AMD and Intel announced that they will combine their x86 CPU optimization work for PyTorch in oneDNN, the open-source deep learning library governed by the Unified Acceleration Foundation (UXL), and have aligned on it as the default open-source library for x86 acceleration inside PyTorch. AMD is folding its ZenDNN optimizations into oneDNN and has joined
via StorageReview
AMD and Intel are working together to bring CPU inference optimizations for PyTorch under the unified oneDNN library.
via Tom’s Hardware
AMD & Intel have expanded their ongoing EAG collaboration by bringing simplified AI development on x86 CPUs using PyTorch. AMD Brings its ZenDNN Optimizations To oneDNN As A One-Stop Shop To Simplify Development on x86 CPUs Using PyTorch Alongside Intel Ever since the formation of the EAG, or the x86 Ecosystem Advisory Group, Intel and AMD have been working closely to advance the x86 segment through deeper collaborations that bring CPU optimizations and various improvements…
via Wccftech