<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Trainium on AI 实战派 · 从技术到赚钱</title><link>https://guijiagi.com/tags/trainium/</link><description>Recent content in Trainium on AI 实战派 · 从技术到赚钱</description><generator>Hugo</generator><language>zh-cn</language><copyright>本站内容采用 CC BY-NC-SA 4.0 国际许可协议授权</copyright><lastBuildDate>Tue, 20 Oct 2026 09:48:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/trainium/index.xml" rel="self" type="application/rss+xml"/><item><title>「亚马逊开源TorchNeuron」PyTorch原生跑Trainium，告别XLA硬编码能撬动CUDA吗？</title><link>https://guijiagi.com/posts/2026-10-20-08-torchneuron-open-source-trainium-pytorch/</link><pubDate>Tue, 20 Oct 2026 09:48:00 +0800</pubDate><guid>https://guijiagi.com/posts/2026-10-20-08-torchneuron-open-source-trainium-pytorch/</guid><description>10月19日PyTorch官方博客宣布开源TorchNeuron原生后端，模型代码与工具不改或少改即可跑在AWS Trainium2与3上，Apache 2.0协议，去NVIDIA化多一条路？</description></item></channel></rss>