https://0hye.com/Recent content on 0hyeHugoenTue, 29 Sep 2026 00:00:00 +0900https://0hye.com/posts/budget-llm-cybersecurity/Tue, 29 Sep 2026 00:00:00 +0900https://0hye.com/posts/budget-llm-cybersecurity/<p>I measured five budget-tier LLMs on two separate things: what they know about security when asked closed-book, and whether they can actually solve a security task as an agent. The short version: <strong>the knowledge scores barely separate them, and the agentic scores separate them a lot.</strong></p> <p>The full, up-to-date numbers live on the <a href="https://0hye.com/benchmarks/budget-llm-cybersecurity/">benchmark page</a>, which I update when a new model comes out. This post explains what was run and how to read it.</p>https://0hye.com/benchmarks/budget-llm-cybersecurity/Mon, 28 Sep 2026 00:00:00 +0900https://0hye.com/benchmarks/budget-llm-cybersecurity/<p>Five similarly priced models ($0.09–0.20 in, $0.36–1.20 out per 1M tokens) on two separate things: what they <strong>know</strong> about security when asked closed-book, and whether they can <strong>do</strong> a security task through a tool-using agent. The write-up is in <a href="https://0hye.com/posts/budget-llm-cybersecurity/">this post</a>.</p>