<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Perf on Nikhil Jain</title><link>http://nikhiljain17.github.io/tags/perf/</link><description>Recent content in Perf on Nikhil Jain</description><generator>Hugo -- 0.146.0</generator><language>en-us</language><lastBuildDate>Tue, 20 Oct 2026 15:35:25 -0700</lastBuildDate><atom:link href="http://nikhiljain17.github.io/tags/perf/index.xml" rel="self" type="application/rss+xml"/><item><title>Reducing Memory Usage for Llama.cpp in the Browser</title><link>http://nikhiljain17.github.io/notes/wllama_streaming_memory_reduction/</link><pubDate>Tue, 20 Oct 2026 15:35:25 -0700</pubDate><guid>http://nikhiljain17.github.io/notes/wllama_streaming_memory_reduction/</guid><description>Streaming weights to reduce memory usage</description></item></channel></rss>