{
  "post": {
    "title": "What an AI agent really costs: cost per completed task",
    "url": "http://bardak.ai/posts/ai-agent-cost-per-task",
    "updated": "2026-11-16T10:26:57.457Z"
  },
  "citation": "What an AI agent really costs: cost per completed task, Ivan Bardak, bardak.ai, http://bardak.ai/posts/ai-agent-cost-per-task",
  "datasets": [
    {
      "key": "swe-bench-verified-with-the-same-agent-cost-per-resolved-issue",
      "title": "SWE-bench Verified with the same agent: cost per resolved issue",
      "type": "benchmarkTable",
      "source": "SWE-bench leaderboard, Bash Only (mini-SWE-agent), as of October 2026",
      "caption": "Cost per resolved issue is my arithmetic: cost per attempt divided by the share resolved. Runs from July 2025 to February 2026, so older models had older prices.",
      "csv": "http://bardak.ai/data/ai-agent-cost-per-task/swe-bench-verified-with-the-same-agent-cost-per-resolved-issue.csv",
      "header": [
        "Model",
        "Resolved",
        "Calls per issue",
        "$ per attempt",
        "$ per resolved issue"
      ],
      "rows": [
        [
          "GPT-5 mini (medium)",
          "59.8%",
          "14.5",
          "0.035",
          "0.059"
        ],
        [
          "MiniMax M2.5 (high)",
          "75.8%",
          "60.5",
          "0.073",
          "0.097"
        ],
        [
          "Gemini 3 Flash (high)",
          "75.8%",
          "56.1",
          "0.356",
          "0.47"
        ],
        [
          "DeepSeek V3.2 (high)",
          "70.0%",
          "88.5",
          "0.448",
          "0.64"
        ],
        [
          "Claude 4.5 Opus (high)",
          "76.8%",
          "32.9",
          "0.754",
          "0.98"
        ],
        [
          "GPT-4.1",
          "39.6%",
          "19.5",
          "0.146",
          "0.37"
        ],
        [
          "GPT-4.1 mini",
          "23.9%",
          "89.2",
          "0.435",
          "1.82"
        ],
        [
          "GPT-4o",
          "21.6%",
          "63.1",
          "1.531",
          "7.08"
        ]
      ]
    },
    {
      "key": "theagentcompany-cost-per-fully-completed-task",
      "title": "TheAgentCompany: cost per fully completed task",
      "type": "chart",
      "source": "TheAgentCompany paper (December 2024) and leaderboard (checked entries, standard environment)",
      "caption": "My arithmetic: the published cost per attempt divided by the share of tasks fully completed (24% to 33%). The benchmark assumes no prompt caching, so real bills would be lower.",
      "csv": "http://bardak.ai/data/ai-agent-cost-per-task/theagentcompany-cost-per-fully-completed-task.csv",
      "header": [
        "agent",
        "Cost per completed task"
      ],
      "rows": [
        [
          "OpenHands + Claude 3.5 Sonnet",
          "26.42"
        ],
        [
          "OpenHands + Claude 3.7 Sonnet",
          "15.41"
        ],
        [
          "OpenHands + GPT-4o",
          "15.00"
        ],
        [
          "OpenHands + Gemini 2.5 Pro",
          "13.97"
        ],
        [
          "OpenHands-Versa + Claude Sonnet 4",
          "4.92"
        ]
      ]
    },
    {
      "key": "list-price-against-real-cost-on-the-same-evaluations",
      "title": "List price against real cost on the same evaluations",
      "type": "benchmarkTable",
      "source": "Artificial Analysis, Intelligence Index v5.0, model pages, accessed 29 October 2026",
      "caption": "Two models with the same $2 and $10 list prices cost $1.27 and $10.02 per task, because they use very different numbers of tokens on the same tasks, and the cheaper one scores higher. Sonnet costs as much per task as Opus at half the price per token. Each model at its highest reasoning setting on Artificial Analysis (High for Gemini 4 Argon), which prices Gemini 4 Argon at its standard $4 and $20 rather than the promotional $2 and $10.",
      "csv": "http://bardak.ai/data/ai-agent-cost-per-task/list-price-against-real-cost-on-the-same-evaluations.csv",
      "header": [
        "Model",
        "In $/1M",
        "Out $/1M",
        "Index score",
        "$ per task"
      ],
      "rows": [
        [
          "Claude Opus 5.5",
          "4.00",
          "20.00",
          "58",
          "10.04"
        ],
        [
          "GPT-6.1 Sol",
          "2.00",
          "10.00",
          "55",
          "1.27"
        ],
        [
          "Claude Sonnet 5.5",
          "2.00",
          "10.00",
          "54",
          "10.02"
        ],
        [
          "Gemini 4 Argon",
          "4.00",
          "20.00",
          "50",
          "7.68"
        ],
        [
          "Claude Haiku 5.5",
          "0.10",
          "0.50",
          "36",
          "0.41"
        ],
        [
          "GPT-6 Luna",
          "0.10",
          "0.50",
          "28",
          "0.15"
        ]
      ]
    },
    {
      "key": "one-agent-three-ways-to-run-it",
      "title": "One agent, three ways to run it",
      "type": "benchmarkTable",
      "source": "",
      "caption": "At 10,000 tasks a month, the difference between the first and last row is about $7,000 a month, before any change of model.",
      "csv": "http://bardak.ai/data/ai-agent-cost-per-task/one-agent-three-ways-to-run-it.csv",
      "header": [
        "Setup",
        "Cost per attempt",
        "Success rate",
        "Cost per completed task"
      ],
      "rows": [
        [
          "No caching",
          "$0.60",
          "60%",
          "$1.00"
        ],
        [
          "70% of input read from cache at a twentieth of the price",
          "$0.33",
          "60%",
          "$0.56"
        ],
        [
          "Same, run as a batch at half price",
          "$0.17",
          "60%",
          "$0.28"
        ]
      ]
    },
    {
      "key": "appraising-the-evidence",
      "title": "Appraising the evidence",
      "type": "benchmarkTable",
      "source": "",
      "caption": "",
      "csv": "http://bardak.ai/data/ai-agent-cost-per-task/appraising-the-evidence.csv",
      "header": [
        "Evidence",
        "Strength",
        "Main limitation"
      ],
      "rows": [
        [
          "SWE-bench and TheAgentCompany costs",
          "Same tasks and setup for every model",
          "Benchmarks, not your workload; costs at the prices of the run date; TheAgentCompany assumes no caching"
        ],
        [
          "Vendor pricing and caching documentation",
          "Authoritative for prices and discounts",
          "Prices change, some are promotional; tokenizers differ, so per-token prices do not compare exactly"
        ],
        [
          "Vendor engineering posts (Anthropic)",
          "Real systems at scale",
          "Written about their own products"
        ],
        [
          "Artificial Analysis",
          "Independent, same evaluations for all models",
          "Its evaluation mix, not your tasks"
        ],
        [
          "My cost tables",
          "Transparent arithmetic",
          "Illustrative numbers"
        ]
      ]
    }
  ]
}