<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title>AI Models on Rost Glukhov | Sistemi IA e infrastruttura</title>
		<link>https://www.glukhov.org/it/tags/ai-models/</link>
		<description>Recent content in AI Models on Rost Glukhov | Sistemi IA e infrastruttura</description>
		<generator>Hugo</generator>
		<language>it</language>
		
		
		
		
			<lastBuildDate>Sun, 15 Nov 2026 15:53:56 +1100</lastBuildDate>
		
			<atom:link href="https://www.glukhov.org/it/tags/ai-models/index.xml" rel="self" type="application/rss+xml" />
			<item>
				<title>Come scegliere un LLM senza censura: abliterated o fine-tuned</title>
				<link>https://www.glukhov.org/it/ai-models/comparisons/choosing-uncensored-abliterated-llm/</link>
				<pubDate>Thu, 12 Nov 2026 00:00:00 +0000</pubDate>
				<guid>https://www.glukhov.org/it/ai-models/comparisons/choosing-uncensored-abliterated-llm/</guid>
				<description>&lt;p&gt;&amp;ldquo;Uncensored&amp;rdquo; copre tre metodi 2026 per modelli aperti che non rifiutano: abliteration, fine-tuning senza rifiuto e politica steerabile. Ognuno modifica capacità diverse; questa guida confronta i costi e seleziona i modelli in base alla VRAM.&lt;/p&gt;</description>
			</item>
	</channel>
</rss>
