<?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>Vllm on Matthew Bonanni</title><link>https://matthewbonanni.github.io/tags/vllm/</link><description>Recent content in Vllm on Matthew Bonanni</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Mon, 09 Nov 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://matthewbonanni.github.io/tags/vllm/index.xml" rel="self" type="application/rss+xml"/><item><title>HiSparse Concurrency Calculator</title><link>https://matthewbonanni.github.io/posts/2026-11-09-hisparse-calculator/</link><pubDate>Mon, 09 Nov 2026 00:00:00 +0000</pubDate><guid>https://matthewbonanni.github.io/posts/2026-11-09-hisparse-calculator/</guid><description>An interactive calculator for estimating how many concurrent long-context requests GLM and DeepSeek sparse-attention models can serve with and without Hybrid HiSparse in vLLM.</description></item></channel></rss>