Google and Meta launch TorchTPU, a native PyTorch backend for TPUs
PyTorch · x · 2026-10-22
PyTorch announced TorchTPU, an open-source native PyTorch backend for Google TPUs developed jointly by Google and Meta in partnership with PyTorch core maintainers, introduced by Bill Jia (Google Cloud) and Zongwei Zhou (Meta) at PyTorchCon 2026.
TorchTPU integrates with Hugging Face Transformers for modeling, TorchTitan for training, and vLLM and SGLang for inference, alongside agentic developer workflows demoed at the conference.
More from Infra
- FlyDSL Kernels Land in PyTorch, Up to 3.01x vLLM Speedups on AMD MI350 — PyTorch · 2026-10-22
- Micron AI memory roadmap: bandwidth moves onto the die, HBM supply tight from 2029 — BenBajarin · 2026-10-22
- Qualcomm's non-phone revenue seen growing 60% next year, driven by auto and embedded — pdamodaran · 2026-10-22
- After PHP advisories, a WebAssembly runtime vendor lays out its security validation plan — MickeySteamboat · 2026-10-22
- Hugging Face: Specializing an SLM's vocabulary can cut inference cost by 77% — Hugging Face Blog · 2026-10-22
- Ollama Runs Open Models on NVIDIA Blackwell via Crusoe, Live in Two Weeks — ollama · 2026-10-22