Google与Meta开源TorchTPU原生PyTorch TPU后端
TorchTPU - A Hybrid Eager/Compiled Architecture for Native PyTorch Execution on TPU
仅依据论文摘要整理；未读取全文，实验条件、证明与基准细节请核对原文。
作者：Claudio Basile, Michael Voznesensky, Gunhyun Park, Ben Wilson, Parker Holloway, Zhanyong Wan, Vlad Belous, Manish Khanna, Paul Ganssle, Kevin Gleason, Andrew Dame, Brendan Folie, Bhavya Bahl, Haifeng Jin, Pedro Goncalves Mokarzel, Arian Arfaian, Luke Boyer, Chi Zhang, Yiling Yuan, Adam Mainz, Aditya Patel, Huangze Tang, Jay Gan, Lizzie Liang, Will Froom, Tushar Agarwal, Karlo Basioli, Theotime Combes, Emilio Cota, Minos Papadopoulos, Tudor Pipernea, Yarong Mu, Qi Zhou, Yaoshiang Ho, Aleksey Vlasenko, Alexander Brauckmann, Ziyin Huang, Jack Oh, Jerry Hsu, Tanvi Bajpai, Jialei Chen, Wenjun Gu, Chintan Shah, Chris Chan, Sayce Falk, Chris Jones, Rob Torres, Kat Ko, Lee Howes, Freya Azad, Andi Gavrilescu, Joe Pamer, Bill Jia
Google与Meta联合推出开源项目TorchTPU，为Google TPU提供原生PyTorch后端，并支持模型直接运行与迁移到TPU。该后端结合eager执行优化、torch.compile集成和原生硬件加速；摘要称其训练与推理性能达到行业竞争力，但未提供具体基准结果。
阶段、条件与复现 · 深读核对
- 新能力对应什么具体任务与最小输入输出？
- 代码、模型、数据、许可与可用入口是否明确？
- 效果、总成本、失效条件与实际工作流如何验证？
这些是阅读核对问题；材料未说明的条件保留未知。请结合上方论文版本、资料范围与原文核验。
来源：arXiv · cs.AI · arxiv.org