A Runaway Prompt Is a Tenant: Fair Scheduling on a Shared Free Endpoint

<p>Anthropic shipped Opus 4.7 on April 16, 2026. Same $5/$25 pricing as 4.6. Same 1M context. Same 128K output ceiling. Different model entirely when you put it in an agentic

<p>My build passed on my laptop. It died on a free server in 40 seconds.</p> <p>Same code. Same lockfile. Same commit hash. So what changed?</p> <p>The answer is boring and

<p>Every code review has the same advice. "Add a retry." "Set a timeout." "Use keep-alive." Nobody measures first. I got tired of guessing. So I built a probe. Connection advice

<p>I keep hitting the same failure pattern. A merge request passes every test. Then the pipeline stalls. Nothing in the diff changed. The free model route I call for a review is

<div class="highlight js-code-highlight"> <pre class="highlight markdown"><code><span class="p">![</span><span class="nv">Architecture Diagram</span><span class="p">](</span><span

<p>Have you ever watched an LLM endpoint return a clean answer that belonged to a different prompt entirely? I spent forty-eight hours blaming sampling noise, temperature, and a

<p>It is 02:14. Your nightly batch window closes at 02:30. The free lane is still running, and the token meter reads zero.</p> <p>Last week that felt like a win. Tonight the

<p>We spend a lot of time testing whether a service can <em>start</em>. We spend far less time testing whether it can <em>exit</em>. A free model can scaffold an HTTP server, and