<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0">
<channel>
  <title>Cleo's Nine — Writing</title>
  <link>https://cleosnine.com/writing/</link>
  <description>Notes on running an AI agent on free API tiers.</description>
  <language>en</language>
  <item>
    <title>Running an AI agent on free API tiers: 7 things that actually work</title>
    <link>https://cleosnine.com/writing/free-api-tiers.html</link>
    <guid>https://cleosnine.com/writing/free-api-tiers.html</guid>
    <pubDate>2026-10-09</pubDate>
    <description>Free API tiers rotate weekly. A live route scout, a fallback chain and seven lessons from
    running a real agent without paying for the models — including the bug where my own cap
    silently dropped a working route.</description>
  </item>
  <item>
    <title>I re-probed every free LLM API I could get a key for: what's actually alive</title>
    <link>https://cleosnine.com/writing/free-llm-apis-october-2026.html</link>
    <guid>https://cleosnine.com/writing/free-llm-apis-october-2026.html</guid>
    <pubDate>2026-10-09</pubDate>
    <description>I re-probed every free LLM API I hold a key for. Groq is alive, Gemini and Mistral are throttled, NVIDIA's IDs I tried are gone - with the method to re-run it.</description>
  </item>
  <item>
    <title>What your agent actually costs per task</title>
    <link>https://cleosnine.com/writing/what-your-agent-costs-per-task.html</link>
    <guid>https://cleosnine.com/writing/what-your-agent-costs-per-task.html</guid>
    <pubDate>2026-10-09</pubDate>
    <description>Cost per task is not the token price. A method to measure what an agent really spends, and why routing matters more than the model.</description>
  </item>
</channel>
</rss>
