NVIDIA Jetson AI 实验室为 JetPack 6.1 提供了预编译的 PyTorch 轮子，可以在 Jetson-AI-Lab 均衡索引中找到。虽然没有批量的特定 2.5.1 轮子，但那里提供的最新版本是 2.6.0 rc1 轮子，它比 2.5.1 新。您可以从以下网址下载它（以及 2.5.0 和 2.4.0）轮子）：
https://pypi.jetson-ai-lab.dev/jp6/cu126/
选择适合您的Python版本的轮子（例如torch-2.6.0-cp310-cp310-linux_aarch64.whl）。如果您特别需要2.5.1，则当前未在首发的源中提供；您需要从源代码构建它或等待官方发布。
参考：Jetson-AI-Lab 目录中提到 JetPack 6.1 的最新预编译版本是 2.6.0rc1，还推出了 2.4.0 和 2.5.0 轮子 [8](https://forums.developer.nvidia.com/t/jetpack-6-1-install-pytorch-2-5-1/318629 "8") [9](https://forums.developer.nvidia.com/t/jetpack-6-1-install-pytorch-2-5-1/318629 "9") 。
Hi,
We provide 2.4.0, 2.5.0, and 2.6.0rc1 for the JetPack 6.1 environment.
https://pypi.jetson-ai-lab.dev/jp6/cu126/torch
For other versions, you can build it from the source.
You can find the building command in the below topic:
PyTorch for JetsonAnnouncements
Below are pre-built PyTorch pip wheel installers for Jetson Nano, TX1/TX2, Xavier, and Orin with JetPack 4.2 and newer. Download one of the PyTorch binaries from below for your version of JetPack, and see the installation instructions to run on your Jetson. These pip wheels are built for ARM aarch64 architecture, so run these commands on your Jetson (not on a host PC). You can also use the containers from jetson-containers. PyTorch pip wheels JetPack 6 ▶ PyTorch v2.3.0 ▶ PyTorch v2.2.0 ▶ PyT...
使用pip下载轮子：
pip3 安装'numpy <2'
$ pip3 install --force torch-2.6.0-cp310-cp310-linux_aarch64.whl
$ pip3 install --force torchvision-0.21.0-cp310-cp310-linux_aarch64.whl
$ pip3 install --force torchaudio-2.6.0-cp310-cp310-linux_aarch64.whl
下面是官方轮子下载
高版本
https://pypi.jetson-ai-lab.dev/jp6/cu126/
低版本
https://forums.developer.nvidia.com/t/pytorch-for-jetson/72048
还有一种自己构建，可能用源码，还不知怎么做
