The source post has been deleted, protected, or is otherwise unavailable.
Why it matters
Max effort scores 68 on the Coding Agent Index versus 63 at xhigh, but costs $14.19 versus $3.33 per task. Engineers can choose effort levels with the cost-performance tradeoff quantified.
Key takeaways · AI-distilled
In Claude Code, Artificial Analysis measured Sonnet 5.5 at 68 on its Coding AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → Index at max effort and 63 at xhigh, at $14.19 versus $3.33 per task, so choosing xhigh cuts cost per task by 77%.
Max effort generated 4.8 times as many output tokenThe chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.Full definition → and took 3.4 times as many turns as xhigh, which explains most of the cost gap.
Max improved all three component benchmarks, with the largest gain on Terminal-Bench 4.0, from 58.1% to 66.2%.
Terms in this piece · Glossary
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
