Anthropic prompt caching cut our RCA cost by 90%

<p>A token allowance is usually treated as a spending budget, which is the wrong mental model for free tiers. The right model is a cache to be managed, because agent workloads

<p><strong>TL;DR: We turned on vLLM's prefix cache for our agent workloads at Nexus Labs and watched TTFT drop from 480ms to 110ms on one tenant and stay exactly the same on

<p>Prompt caching is not a setting you enable. It is a property your prompt either has or does not have, and one timestamp in the wrong place removes it from every request you

<p>GPT-5.4 shipped on March 5. Seven weeks later, on April 23, 2026, OpenAI released GPT-5.5 — and the pace alone tells you something about where this race is headed. This isn't

<h1> Prompt Caching: First vs Repeat Request Costs </h1> <p>Measure prompt-cache cost with a controlled request series: the first request creates or prepares a cached prefix,

<p>LLM API costs add up fast when you're running a production application. If your system prompt is 2,000 tokens and you're processing 10,000 requests a day, you're paying for

<p>The price page hasn't moved. Five dollars per million input. Twenty-five per million output. Identical to Opus 4.6.</p> <p>Run the same Claude Code call on both, though, and

<blockquote> <p>Skip the theory rabbit holes. This is the caching knowledge that shows up in system design interviews, code reviews, and the 2 AM production incidents nobody