Token cost (cost per task)
The dollar cost of a model run, priced per million input and output tokens; agentic coding multiplies it through long loops.
Claude Opus 4 launched at $15/$75 per million input/output tokens, Sonnet 4 at $3/$15. A single agent task can consume hundreds of thousands of tokens, so cost per resolved task varies by 100× across models with similar scores (Aider: 88% for $29 vs 74% for $1.30 per full run). Leaderboards increasingly report cost, and SWE-bench Pro distinguishes capped and uncapped budgets.