← This week · All arcs

PyTorch hardware backends

3 stories · Oct 21, 2026 to Oct 24, 2026
The stories of one ongoing situation, in the order they were first reported. Which stories belong together is a model's grouping; each story's sources are the reports that carried it.

What happened

TechCrunch reported on October 19, 2026, that AWS had released TorchNeuron.1 According to that report, PyTorch AI apps could move onto Trainium with one config line and no code rewrite.1 Hugging Face reported the same day that TorchNeuron, a PyTorch backend for AWS Trainium, had entered public beta.2 Hugging Face said Transformers, Diffusers, Accelerate, PEFT and TRL ran natively on it.2

Amazon's code and Google's release

TechCrunch reported on October 20 that Amazon was submitting code to decouple PyTorch from Nvidia hardware.3 TechCrunch also reported that day that Google had made TorchTPU, its open source PyTorch version, available.4 According to TechCrunch, apps could run on TPUs with a simple configuration change and no rewriting.4

Hugging Face reported on October 20 that it was adding native TPU support to Transformers, Diffusers and Kernels through TorchTPU.5 Hugging Face said Optimum TPU was no longer required to run its models on TPUs.5

PyTorch 2.15 for Apple silicon

Also on October 20, the Hugging Face PyTorch team published a release post on PyTorch 2.15, with a release candidate available.6 The post covered FP8 on MPS, faster MoE and linear decoding, Metal kernels and memory fixes.6 Latent Space reported on October 21 that PyTorch 2.15 replaces MPSGraph paths with Metal kernels, with 5-6.5x faster MoE decoding.7 Latent Space also reported Qwen3.6-35B-A3B running at 61.3 tokens/s on M3 Ultra.8

Latest reported

Latent Space reported on October 22 that Google had open-sourced TorchTPU for PyTorch training and serving on TPUs.9 Latent Space said SGLang and vLLM plugins add caching, speculative decoding and disaggregated serving.10

Written by a model (opus) on Nov 6, 2026 from what this page showed that day, and checked sentence by sentence against it; 15 sentences shown. Each source line is the record's summary of a report.
Sources for the overview
  1. TechCrunch: AWS releases TorchNeuron Oct 19, 2026 · TechCrunch
  2. Hugging Face: TorchNeuron launches with Hugging Face library support Oct 19, 2026 · Hugging Face
  3. TechCrunch: Amazon releases TorchNeuron Oct 20, 2026 · TechCrunch
  4. TechCrunch: Google releases TorchTPU Oct 20, 2026 · TechCrunch
  5. Hugging Face: TorchTPU brings native TPU support to Hugging Face Oct 20, 2026 · Hugging Face
  6. Hugging Face: PyTorch 2.15 brings MPS FP8 and faster decoding Oct 20, 2026 · Hugging Face
  7. Latent Space: PyTorch 2.15 Metal kernels for Apple silicon Oct 21, 2026 · Latent Space
  8. Latent Space: PyTorch 2.15 Metal kernel gains Oct 21, 2026 · Latent Space
  9. Latent Space: Google open-sources TorchTPU Oct 22, 2026 · Latent Space
  10. Latent Space: Google open-sources TorchTPU Oct 22, 2026 · Latent Space

The stories

Built 2026-11-21 06:50 UTC from the news record.