TREE NEWS update: AMD and Intel have agreed to make oneDNN the default open-source library for x86 acceleration in PyTorch. AMD has also joined the UXL Foundation as a steering member to advance the oneDNN PyTorch work. The collaboration aims to simplify AI development on x86 CPUs.
AMD and Intel Partner to Make oneDNN Default x86 Backend for PyTorch
AI take
The significance here is less about a single library and more about two longtime rivals converging on shared x86 infrastructure: making oneDNN the default PyTorch backend lowers fragmentation for developers who currently navigate vendor-specific acceleration paths. AMD's steering membership in the UXL Foundation suggests this is an institutional commitment rather than a one-off patch. Whether the arrangement actually simplifies AI development on x86 CPUs — and whether the ecosystem consolidates around it rather than splintering into competing defaults — is the open question worth watching.
Generated by AI for reference only.
Share on WeChat
Open WeChat → Scan → then tap "…" to send to a chat or Moments.
Tap "…" in the top-right corner to send to a chat or share to Moments.
Related News
17m ago
Altman Floats Merging Top AI Labs Into One Body for Safety
23m ago
HSBC CEO warns not all data centers are good bets amid AI boom
1h ago
Google launches new Gemini tools for small business owners
1h ago
France’s Mistral Releases Voxtral 3 Speech-to-Text and Text-to-Speech Model
2h ago
Judge Rejects Trump Administration Bid to Dismiss NAACP Suit Against Musk’s xAI
2h ago
Dema Technology, Ant Group’s Lingbo Form Strategic Partnership on Embodied AI