<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: davada</title><link>https://news.ycombinator.com/user?id=davada</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 19 Nov 2026 16:12:06 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=davada" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[Keystroke Recognition Using WiFi Signals [pdf]]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.sigmobile.org/mobicom/2015/papers/p90-aliA.pdf">https://www.sigmobile.org/mobicom/2015/papers/p90-aliA.pdf</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=50403956">https://news.ycombinator.com/item?id=50403956</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Sat, 07 Nov 2026 12:25:15 +0000</pubDate><link>https://www.sigmobile.org/mobicom/2015/papers/p90-aliA.pdf</link><dc:creator>davada</dc:creator><comments>https://news.ycombinator.com/item?id=50403956</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50403956</guid></item><item><title><![CDATA[New comment by davada in "Run QWEN3.8 27B on 16gb Nvidia GPUs"]]></title><description><![CDATA[
<p>Btw, using llama.cpp you can achieve 55 - 45 token/second for processing/generation. if you use a qwen3.8-27B (IQ4_XS) verison, with decent quality in reasoning for coding/tool usage (with a 16GB nvidia).<p>I think now most of the struggle is getting a 32B-27B llm to work on 16GB/12GB card, because they are at least affordable/accessible for the time being compared to higher end models.<p>Recently the nvidia RTX 5090 32GB has reached price range of 7500 dollars (despite MSRP being around 2000 dollars when it was first launched). Crazy times.</p>
]]></description><pubDate>Sun, 20 Sep 2026 03:37:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=49772325</link><dc:creator>davada</dc:creator><comments>https://news.ycombinator.com/item?id=49772325</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49772325</guid></item><item><title><![CDATA[New comment by davada in "Run QWEN3.8 27B on 16gb Nvidia GPUs"]]></title><description><![CDATA[
<p>Its not so secretive, you can see the parameters they are using (arguments, and inference engine). I understand your point, but the only thing unusual/uncommon about it is the jargon used for LLM/AI systems.</p>
]]></description><pubDate>Sun, 20 Sep 2026 03:34:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=49772311</link><dc:creator>davada</dc:creator><comments>https://news.ycombinator.com/item?id=49772311</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49772311</guid></item></channel></rss>