---
type: "raw_signal"
id: "mps-in-pytorch-2-15-841c7d"
slug: "mps-in-pytorch-2-15-841c7d"
title: "MPS in PyTorch 2.15"
description: "Hugging Face’s PyTorch team describes MPS backend changes in PyTorch 2.15 for Apple silicon."
retrieval_nugget: "Hugging Face’s PyTorch team describes MPS backend changes in PyTorch 2.15 for Apple silicon."
observed_at: "2026-10-20"
record_date: "2026-10-20"
date_kind: "observed_at"
why_it_matters: "Hugging Face’s PyTorch team describes MPS backend changes in PyTorch 2.15 for Apple silicon. The release adds MPS support for `torch.float8_e4m3fn` tensors and scaled matrix multiplication, and introduces Metal implementations for operations including grouped matrix multiplication and softmax. In the authors’ reported tests, Qwen3.6-35B-A3B decoding on an M3 Ultra with `torch.compile` increased fr"
novelty: "notable"
verification_level: "source-linked"
signal_type: "release"
source_platform: "huggingface.co"
topics: []
entities: ["Hugging Face","PyTorch","Apple MPS","Qwen3.6-35B-A3B","Wan2.2"]
related_patterns: []
source_url: "https://huggingface.co/blog/mps-pytorch-2-15"
source_urls: ["https://huggingface.co/blog/mps-pytorch-2-15"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "signal:mps-in-pytorch-2-15-841c7d"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/signals/mps-in-pytorch-2-15-841c7d/","markdown":"https://newruntime.com/signals/mps-in-pytorch-2-15-841c7d.md","json":"https://newruntime.com/signals/mps-in-pytorch-2-15-841c7d.json"}
---

# MPS in PyTorch 2.15

## Retrieval answer

Hugging Face’s PyTorch team describes MPS backend changes in PyTorch 2.15 for Apple silicon.

Hugging Face’s PyTorch team describes MPS backend changes in PyTorch 2.15 for Apple silicon. The release adds MPS support for `torch.float8_e4m3fn` tensors and scaled matrix multiplication, and introduces Metal implementations for operations including grouped matrix multiplication and softmax. In the authors’ reported tests, Qwen3.6-35B-A3B decoding on an M3 Ultra with `torch.compile` increased from 9.4 to 61.3 tokens per second compared with PyTorch 2.14. For Wan2.2 tiled VAE decoding, reported peak total driver memory fell from about 270 GiB to 8 GiB. These are workload-specific measurements, not general guarantees. The article says PyTorch 2.15 requires Python 3.11 or later and MPS requires macOS 15 or later. Source: https://huggingface.co/blog/mps-pytorch-2-15

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Raw Signals](https://newruntime.com/signals.json): Open every source-linked raw signal.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)