<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Gated-Delta-Networks on Noureddine RAMDI</title><link>https://ramdi.fr/tags/gated-delta-networks/</link><description>Recent content in Gated-Delta-Networks on Noureddine RAMDI</description><generator>Hugo</generator><language>en</language><lastBuildDate>Sat, 23 May 2026 20:41:27 +0000</lastBuildDate><atom:link href="https://ramdi.fr/tags/gated-delta-networks/index.xml" rel="self" type="application/rss+xml"/><item><title>Alibaba's Qwen3.6: Efficient large-scale LLMs with gated delta networks and sparse MoE</title><link>https://ramdi.fr/github-stars/alibaba-s-qwen3-6-efficient-large-scale-llms-with-gated-delta-networks-and-sparse-moe/</link><pubDate>Tue, 05 May 2026 13:37:39 +0000</pubDate><guid>https://ramdi.fr/github-stars/alibaba-s-qwen3-6-efficient-large-scale-llms-with-gated-delta-networks-and-sparse-moe/</guid><description>Qwen3.6 from Alibaba uses gated delta networks and sparse Mixture-of-Experts to achieve near-397B parameter model performance with only 3B active parameters, supporting 201 languages and 262k context length.</description></item></channel></rss>