Google and Meta open-source TorchTPU: native PyTorch backend for TPUs
PyTorch · x · 2026-10-22
At PyTorchCon 2026, Bill Jia (Google Cloud) and Zongwei Zhou (Meta) introduced TorchTPU, an open-source native PyTorch backend for Google TPUs, built in close partnership with PyTorch core maintainers.
- Integrates with Hugging Face Transformers for modeling, TorchTitan for training, and vLLM/SGLang for serving
- Demonstrated migrating complex model workloads from GPUs to TPUs via long-horizon agentic workflows
- Also showed agents used for performance optimization: quantization, custom kernel generation, and advanced sharding strategies
More from Infra
- NVIDIA keynote: silicon photonics CPO switches hit 102.4 Tb/s, Vera Rubin architecture unveiled — thealexbanks · 2026-10-22
- PyTorch Foundation Adds Nine Members, Nebius, Nscale and Shopify Join as Platinum — PyTorch · 2026-10-22
- GPU startup Flappy Airplanes raises $1B Series A co-led by Index and Lux — vitaliychiley · 2026-10-22
- Crusoe unveils Tinker API-compatible LLM post-training platform at PyTorch Conference — PyTorch · 2026-10-22
- Open PR optimizes DeepSeek-V4.1 inference on RTX PRO 6000 Blackwell — AlpinDale · 2026-10-22
- vLLM adds day-0 support for Flint: 500 TPS on a single GB200 — _rockt · 2026-10-22