Skip to content
K = 0.731
Original
ProductsHugging Face BlogFrom the source·· 18 hours agoPicks

Hugging Face adds native TPU support with TorchTPU

Hugging Face ❤️ TorchTPU: TPUs, Natively

Who
Hugging Face
What
introduces TorchTPU, a native PyTorch backend for running Hugging Face libraries on TPUs
Quote
With TorchTPU , a native PyTorch backend, TPU is now just another device.

What happened

Hugging Face announces native TPU support across Transformers, Diffusers, and Kernels through TorchTPU, a native PyTorch backend. Models can use device_map="tpu" or pipe.to("tpu"), with optional compilation and distributed execution; Optimum TPU is no longer required for Transformers and Diffusers. Hugging Face reports that 96.5% of tests running on TPU pass, with 19 of 34 reported TorchTPU issues already fixed.

Why it matters

Native TPU support lets PyTorch users keep familiar Hugging Face workflows, reducing the software changes previously required to move models onto Google’s accelerators.

The story so far

Hugging Face has added native TPU support across Transformers, Diffusers, and Kernels through TorchTPU, a native PyTorch backend, removing the need for Optimum TPU when running Transformers and Diffusers models on TPUs. Models can target the hardware with device_map="tpu" or pipe.to("tpu"), with optional compilation and distributed execution. Hugging Face reports that 96.5% of tests running on TPU pass and that 19 of 34 reported TorchTPU issues have already been fixed.

Hugging FaceOpen sourceDeployment engineeringProduct updates#Product update#Deployment and engineering#Open-source ecosystem#Hugging Face#Google

Keep reading

Latest daily · Issue 27TorchNeuron brings native PyTorch support for Hugging Face models on AWS Trainium

Trending now

  1. 1DeepSeek Open-Sources DeepInfer and Companion Components, With Model and Hardware Limits
  2. 2Mistral Releases Voxtral 3 Speech Models and LiveKit-Based Voice Agent SDK
  3. 3Perplexity Adds Fully Local Mac Operation to Computer
  4. 4OpenAI Introduces Coding Evaluations and Reports Gains in Low-Resource Programming Languages

Tomorrow at 08:00, a briefing like this arrives in your inbox.

The same kind of briefing as the one above: a few stories worth reading. Free. Mail starts after you confirm, and every letter has a one-click unsubscribe. Read a past issue · 2026-10-20