arXiv 人工智能· 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·· 4 小时前精选AI 评分66
Google 与 Meta 联合开源 TorchTPU：面向 TPU 的原生 PyTorch 混合执行架构
TorchTPU - A Hybrid Eager/Compiled Architecture for Native PyTorch Execution on TPU
AI 导读
Google 与 Meta 联合开源 TorchTPU，为 Google TPU 提供原生的 PyTorch 后端支持。该方案通过动态图执行优化、torch.compile 集成与原生硬件加速，实现开箱即用的模型运行与无缝 TPU 迁移。系统在保持原生 PyTorch 编程体验的同时，在模型推理与训练任务中提供具备行业竞争力的性能表现。
推荐理由
Google 与 Meta 联合开源了面向 TPU 的原生 PyTorch 后端，为开发者在大规模训练和推理中迁移硬件架构提供了直接参考。
来源：arXiv 人工智能 · arxiv.org