<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Yestino - The Signal · Benchmarking Qwen3.8</title><link>https://yestino.com/entities/benchmarking-qwen3-8-327137</link><description>Every event involving Benchmarking Qwen3.8</description><language>en</language><atom:link href="https://yestino.com/entities/benchmarking-qwen3-8-327137/feed.xml" rel="self" type="application/rss+xml"/><item><title>Benchmarking Qwen3.8 27B quantizations: 4-bit holds up, 1-bit collapses</title><link>https://yestino.com/events/benchmarking-qwen3-8-27b-quantizations-4-bit-holds-up-1-bit-99c839</link><guid isPermaLink="true">https://yestino.com/events/benchmarking-qwen3-8-27b-quantizations-4-bit-holds-up-1-bit-99c839</guid><pubDate>Wed, 26 Aug 2026 17:19:43 GMT</pubDate><description>I test Unsloth GGUFs of Qwen3.8 27B (Q4_K_M, UD-Q2_K_XL, UD-IQ1_S) with llama.cpp on GPQA Diamond, IFBench, Terminal-Bench 2.1.
Sources: Hacker News</description></item></channel></rss>