---
id: 20261021-T0-02
title: "TorchTPU：让PyTorch原生跑在TPU上的混合Eager/编译架构"
title_en: "TorchTPU: A Hybrid Eager/Compiled Architecture for Native PyTorch on TPU"
url: https://ai.daily.yangsir.net/daily/20261021-T0-02
issue_date: 2026-10-21
publish_date: 2026-10-20T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2610.24161
---

# TorchTPU：让PyTorch原生跑在TPU上的混合Eager/编译架构

arXiv新论文介绍TorchTPU，这是Google与Meta合作的开源项目，为Google TPU提供原生的PyTorch后端。它支持开箱即用的模型执行和TPU迁移，并提供有行业竞争力的推理性能。方案采用混合的Eager/编译架构，兼顾灵活性。对使用PyTorch的团队来说，这意味着迁移到TPU时可能不再需要大量改写代码。

## English Version

**TorchTPU: A Hybrid Eager/Compiled Architecture for Native PyTorch on TPU**

A new arXiv paper introduces TorchTPU, an open-source collaboration between Google and Meta that provides a native PyTorch backend for Google TPUs. It enables out-of-the-box model execution and TPU migration while delivering industry-competitive inference performance, using a hybrid eager/compiled architecture that balances flexibility and speed. For PyTorch teams, it could mean migrating to TPU without extensive code rewrites.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2610.24161)

**详情页**：https://ai.daily.yangsir.net/daily/20261021-T0-02

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*