Hugging Face Blog·· 14 hours agoFeaturedAI Score67
Hugging Face details PyTorch 2.15 MPS changes: FP8, faster MoE decoding and Metal kernels
MPS in PyTorch 2.15
AI summary
Hugging Face's PyTorch team outlined what PyTorch 2.15 brings to the MPS backend on Apple silicon.
Why it matters
The post pairs each MPS change with measured speedups and memory numbers, so readers can weigh the upgrade against their own Mac workloads.
Source: Hugging Face Blog · huggingface.co