<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>devtake.dev — #huawei</title><description>Articles tagged huawei on devtake.dev.</description><link>https://devtake.dev/</link><language>en-us</language><item><title>China built the world&apos;s fastest supercomputer without a single US chip</title><link>https://devtake.dev/article/china-fastest-supercomputer-top500/</link><guid isPermaLink="true">https://devtake.dev/article/china-fastest-supercomputer-top500/</guid><description>A Chinese system called LineShine debuted at No. 1 on the TOP500 at ISC&apos;26, running entirely on domestic CPUs co-designed with Huawei. Here&apos;s what that means.</description><pubDate>Mon, 29 Jun 2026 10:00:00 GMT</pubDate><category>hardware</category><category>ai-chips</category><category>semiconductor</category><category>china</category><category>huawei</category><category>nvidia</category><category>export-controls</category><category>ai-infrastructure</category><author>hiro-tanaka</author></item><item><title>GLM-5.2 was trained on Huawei chips, not Nvidia. The open weights beat GPT-5.5 on coding.</title><link>https://devtake.dev/article/glm-5-2-open-weights-huawei/</link><guid isPermaLink="true">https://devtake.dev/article/glm-5-2-open-weights-huawei/</guid><description>Zhipu AI&apos;s GLM-5.2 is a free-to-download model trained without Nvidia silicon. Here&apos;s what the benchmarks claim and why developers should care.</description><pubDate>Tue, 23 Jun 2026 09:15:00 GMT</pubDate><category>ai</category><category>ai-models</category><category>open-weights</category><category>llm</category><category>moe</category><category>ai-chips</category><category>benchmarks</category><category>china</category><category>huawei</category><author>dieter-morelli</author></item></channel></rss>