Intel and AMD unite PyTorch CPU optimizations in oneDNN for a common x86 path
PyTorch · x · 2026-10-23
Intel and AMD are consolidating their x86 CPU optimization work for PyTorch into oneDNN, the open-source library governed by the UXL Foundation.
Key points:
- Following a recommendation from the x86 Ecosystem Advisory Group, oneDNN becomes the default open-source x86 acceleration library in PyTorch; AMD is folding its ZenDNN optimizations in and has joined UXL as a Steering Member
- Developers get one consistent software path to build, maintain, and deploy PyTorch apps across Intel and AMD CPUs, reducing x86 AI ecosystem fragmentation
- The two are also collaborating on AI Compute Extensions, a next-gen x86 ISA for matrix multiplication, with future PyTorch integration via oneDNN updates
More from Infra
- Researcher: Putting a $200K GPU Server Purchase Order in a Paper Envelope — HildeKuehne · 2026-10-23
- Google and Meta open-source TorchTPU, a native PyTorch backend for TPUs — PyTorch · 2026-10-23
- Memory-bound at batch 1, compute-bound at batch N: how GPUs actually run LLMs — techNmak · 2026-10-23
- Intel demos AI Playground Home Agent: control local PyTorch compute from your phone via Slack — PyTorch · 2026-10-23
- TorchTPU and TorchNeuron Backends Land as PyTorch Becomes AI's Hardware Battleground — PyTorch · 2026-10-23
- TokenGate e2e-learns to drop tokens on the fly, opening a new frontier for LLM prefill — CShorten30 · 2026-10-23