{
  "name": "GENZ TECH AI Coding Leaderboard — History",
  "description": "Longitudinal record of verified AI coding-model benchmark scores and API prices, reconstructed from every revision of our leaderboard since tracking began. Includes price-per-solved-task, a derived metric published nowhere else.",
  "url": "https://genztech.blog/ai-coding-leaderboard/",
  "license": "https://creativecommons.org/licenses/by/4.0/",
  "attribution": "Data by GENZ TECH (https://genztech.blog/ai-coding-leaderboard/). Free to cite with attribution + a link.",
  "metric": "SWE-bench Verified",
  "firstTracked": "2026-07-02",
  "lastTracked": "2026-07-17",
  "snapshotCount": 7,
  "note": "Only models with a rank (i.e. a score confirmed against a primary source) are included. Prices are USD per 1M tokens as published by the vendor at that date. pricePerSolvedTask = inputPricePer1M / (sweBenchVerified / 100).",
  "snapshots": [
    {
      "date": "2026-07-02",
      "commit": "21f477c18",
      "updated": "2026-07-02",
      "rankedCount": 2,
      "leader": "Claude Opus 4.8",
      "models": [
        {
          "name": "Claude Opus 4.8",
          "maker": "Anthropic",
          "rank": 1,
          "open": false,
          "sweBenchVerified": "~86%",
          "sweBenchPro": "69.2%",
          "inputPricePer1M": "$5",
          "outputPricePer1M": "$25",
          "pricePerSolvedTask": 5.81
        },
        {
          "name": "Claude Sonnet 5",
          "maker": "Anthropic",
          "rank": 2,
          "open": false,
          "sweBenchVerified": "85.2%",
          "sweBenchPro": "63.2%",
          "inputPricePer1M": "$2",
          "outputPricePer1M": "$10",
          "pricePerSolvedTask": 2.35
        }
      ]
    },
    {
      "date": "2026-07-02",
      "commit": "3c079a9bc",
      "updated": "2026-07-02",
      "rankedCount": 7,
      "leader": "Claude Opus 4.8",
      "models": [
        {
          "name": "Claude Opus 4.8",
          "maker": "Anthropic",
          "rank": 1,
          "open": false,
          "sweBenchVerified": "~86%",
          "sweBenchPro": "69.2%",
          "inputPricePer1M": "$5",
          "outputPricePer1M": "$25",
          "pricePerSolvedTask": 5.81
        },
        {
          "name": "Claude Sonnet 5",
          "maker": "Anthropic",
          "rank": 2,
          "open": false,
          "sweBenchVerified": "85.2%",
          "sweBenchPro": "63.2%",
          "inputPricePer1M": "$2",
          "outputPricePer1M": "$10",
          "pricePerSolvedTask": 2.35
        },
        {
          "name": "GPT-5.5",
          "maker": "OpenAI",
          "rank": 3,
          "open": false,
          "sweBenchVerified": "82.6%",
          "sweBenchPro": "58.6%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "DeepSeek V4 Pro",
          "maker": "DeepSeek",
          "rank": 4,
          "open": true,
          "sweBenchVerified": "80.6%",
          "sweBenchPro": "55.4%",
          "inputPricePer1M": "$0.435",
          "outputPricePer1M": "$0.87",
          "pricePerSolvedTask": 0.54
        },
        {
          "name": "Gemini 3.1 Pro",
          "maker": "Google DeepMind",
          "rank": 5,
          "open": false,
          "sweBenchVerified": "80.6%",
          "sweBenchPro": "54.2%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "Kimi K2.6",
          "maker": "Moonshot AI",
          "rank": 6,
          "open": true,
          "sweBenchVerified": "80.2%",
          "sweBenchPro": "58.6%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "Gemini 3.5 Flash",
          "maker": "Google DeepMind",
          "rank": 7,
          "open": false,
          "sweBenchVerified": "78.8%",
          "sweBenchPro": null,
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        }
      ]
    },
    {
      "date": "2026-07-02",
      "commit": "6cec075c3",
      "updated": "2026-07-02",
      "rankedCount": 9,
      "leader": "Claude Opus 4.8",
      "models": [
        {
          "name": "Claude Opus 4.8",
          "maker": "Anthropic",
          "rank": 1,
          "open": false,
          "sweBenchVerified": "~86%",
          "sweBenchPro": "69.2%",
          "inputPricePer1M": "$5",
          "outputPricePer1M": "$25",
          "pricePerSolvedTask": 5.81
        },
        {
          "name": "Claude Sonnet 5",
          "maker": "Anthropic",
          "rank": 2,
          "open": false,
          "sweBenchVerified": "85.2%",
          "sweBenchPro": "63.2%",
          "inputPricePer1M": "$2",
          "outputPricePer1M": "$10",
          "pricePerSolvedTask": 2.35
        },
        {
          "name": "GPT-5.5",
          "maker": "OpenAI",
          "rank": 3,
          "open": false,
          "sweBenchVerified": "82.6%",
          "sweBenchPro": "58.6%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "DeepSeek V4 Pro",
          "maker": "DeepSeek",
          "rank": 4,
          "open": true,
          "sweBenchVerified": "80.6%",
          "sweBenchPro": "55.4%",
          "inputPricePer1M": "$0.435",
          "outputPricePer1M": "$0.87",
          "pricePerSolvedTask": 0.54
        },
        {
          "name": "Gemini 3.1 Pro",
          "maker": "Google DeepMind",
          "rank": 5,
          "open": false,
          "sweBenchVerified": "80.6%",
          "sweBenchPro": "54.2%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "MiniMax M3",
          "maker": "MiniMax",
          "rank": 6,
          "open": true,
          "sweBenchVerified": "80.5%",
          "sweBenchPro": "59.0%",
          "inputPricePer1M": "$0.60",
          "outputPricePer1M": null,
          "pricePerSolvedTask": 0.75
        },
        {
          "name": "Qwen3.7 Max",
          "maker": "Alibaba",
          "rank": 7,
          "open": false,
          "sweBenchVerified": "80.4%",
          "sweBenchPro": "60.6%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "Kimi K2.6",
          "maker": "Moonshot AI",
          "rank": 8,
          "open": true,
          "sweBenchVerified": "80.2%",
          "sweBenchPro": "58.6%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "Gemini 3.5 Flash",
          "maker": "Google DeepMind",
          "rank": 9,
          "open": false,
          "sweBenchVerified": "78.8%",
          "sweBenchPro": null,
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        }
      ]
    },
    {
      "date": "2026-07-03",
      "commit": "c75578904",
      "updated": "2026-07-03",
      "rankedCount": 10,
      "leader": "Claude Fable 5",
      "models": [
        {
          "name": "Claude Fable 5",
          "maker": "Anthropic",
          "rank": 1,
          "open": false,
          "sweBenchVerified": null,
          "sweBenchPro": "80.3%",
          "inputPricePer1M": "$10",
          "outputPricePer1M": "$50",
          "pricePerSolvedTask": null
        },
        {
          "name": "Claude Opus 4.8",
          "maker": "Anthropic",
          "rank": 2,
          "open": false,
          "sweBenchVerified": "~86%",
          "sweBenchPro": "69.2%",
          "inputPricePer1M": "$5",
          "outputPricePer1M": "$25",
          "pricePerSolvedTask": 5.81
        },
        {
          "name": "Claude Sonnet 5",
          "maker": "Anthropic",
          "rank": 3,
          "open": false,
          "sweBenchVerified": "85.2%",
          "sweBenchPro": "63.2%",
          "inputPricePer1M": "$2",
          "outputPricePer1M": "$10",
          "pricePerSolvedTask": 2.35
        },
        {
          "name": "GPT-5.5",
          "maker": "OpenAI",
          "rank": 4,
          "open": false,
          "sweBenchVerified": "82.6%",
          "sweBenchPro": "58.6%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "DeepSeek V4 Pro",
          "maker": "DeepSeek",
          "rank": 5,
          "open": true,
          "sweBenchVerified": "80.6%",
          "sweBenchPro": "55.4%",
          "inputPricePer1M": "$0.435",
          "outputPricePer1M": "$0.87",
          "pricePerSolvedTask": 0.54
        },
        {
          "name": "Gemini 3.1 Pro",
          "maker": "Google DeepMind",
          "rank": 6,
          "open": false,
          "sweBenchVerified": "80.6%",
          "sweBenchPro": "54.2%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "MiniMax M3",
          "maker": "MiniMax",
          "rank": 7,
          "open": true,
          "sweBenchVerified": "80.5%",
          "sweBenchPro": "59.0%",
          "inputPricePer1M": "$0.60",
          "outputPricePer1M": null,
          "pricePerSolvedTask": 0.75
        },
        {
          "name": "Qwen3.7 Max",
          "maker": "Alibaba",
          "rank": 8,
          "open": false,
          "sweBenchVerified": "80.4%",
          "sweBenchPro": "60.6%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "Kimi K2.6",
          "maker": "Moonshot AI",
          "rank": 9,
          "open": true,
          "sweBenchVerified": "80.2%",
          "sweBenchPro": "58.6%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "Gemini 3.5 Flash",
          "maker": "Google DeepMind",
          "rank": 10,
          "open": false,
          "sweBenchVerified": "78.8%",
          "sweBenchPro": null,
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        }
      ]
    },
    {
      "date": "2026-07-03",
      "commit": "34e0208ed",
      "updated": "2026-07-03",
      "rankedCount": 10,
      "leader": "Claude Fable 5",
      "models": [
        {
          "name": "Claude Fable 5",
          "maker": "Anthropic",
          "rank": 1,
          "open": false,
          "sweBenchVerified": "95.0%",
          "sweBenchPro": "80.3%",
          "inputPricePer1M": "$10",
          "outputPricePer1M": "$50",
          "pricePerSolvedTask": 10.53
        },
        {
          "name": "Claude Opus 4.8",
          "maker": "Anthropic",
          "rank": 2,
          "open": false,
          "sweBenchVerified": "~86%",
          "sweBenchPro": "69.2%",
          "inputPricePer1M": "$5",
          "outputPricePer1M": "$25",
          "pricePerSolvedTask": 5.81
        },
        {
          "name": "Claude Sonnet 5",
          "maker": "Anthropic",
          "rank": 3,
          "open": false,
          "sweBenchVerified": "85.2%",
          "sweBenchPro": "63.2%",
          "inputPricePer1M": "$2",
          "outputPricePer1M": "$10",
          "pricePerSolvedTask": 2.35
        },
        {
          "name": "GPT-5.5",
          "maker": "OpenAI",
          "rank": 4,
          "open": false,
          "sweBenchVerified": "82.6%",
          "sweBenchPro": "58.6%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "DeepSeek V4 Pro",
          "maker": "DeepSeek",
          "rank": 5,
          "open": true,
          "sweBenchVerified": "80.6%",
          "sweBenchPro": "55.4%",
          "inputPricePer1M": "$0.435",
          "outputPricePer1M": "$0.87",
          "pricePerSolvedTask": 0.54
        },
        {
          "name": "Gemini 3.1 Pro",
          "maker": "Google DeepMind",
          "rank": 6,
          "open": false,
          "sweBenchVerified": "80.6%",
          "sweBenchPro": "54.2%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "MiniMax M3",
          "maker": "MiniMax",
          "rank": 7,
          "open": true,
          "sweBenchVerified": "80.5%",
          "sweBenchPro": "59.0%",
          "inputPricePer1M": "$0.60",
          "outputPricePer1M": null,
          "pricePerSolvedTask": 0.75
        },
        {
          "name": "Qwen3.7 Max",
          "maker": "Alibaba",
          "rank": 8,
          "open": false,
          "sweBenchVerified": "80.4%",
          "sweBenchPro": "60.6%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "Kimi K2.6",
          "maker": "Moonshot AI",
          "rank": 9,
          "open": true,
          "sweBenchVerified": "80.2%",
          "sweBenchPro": "58.6%",
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        },
        {
          "name": "Gemini 3.5 Flash",
          "maker": "Google DeepMind",
          "rank": 10,
          "open": false,
          "sweBenchVerified": "78.8%",
          "sweBenchPro": null,
          "inputPricePer1M": null,
          "outputPricePer1M": null,
          "pricePerSolvedTask": null
        }
      ]
    },
    {
      "date": "2026-07-03",
      "commit": "a1cce79ae",
      "updated": "2026-07-03",
      "rankedCount": 10,
      "leader": "Claude Fable 5",
      "models": [
        {
          "name": "Claude Fable 5",
          "maker": "Anthropic",
          "rank": 1,
          "open": false,
          "sweBenchVerified": "95.0%",
          "sweBenchPro": "80.3%",
          "inputPricePer1M": "$10",
          "outputPricePer1M": "$50",
          "pricePerSolvedTask": 10.53
        },
        {
          "name": "Claude Opus 4.8",
          "maker": "Anthropic",
          "rank": 2,
          "open": false,
          "sweBenchVerified": "~86%",
          "sweBenchPro": "69.2%",
          "inputPricePer1M": "$5",
          "outputPricePer1M": "$25",
          "pricePerSolvedTask": 5.81
        },
        {
          "name": "Claude Sonnet 5",
          "maker": "Anthropic",
          "rank": 3,
          "open": false,
          "sweBenchVerified": "85.2%",
          "sweBenchPro": "63.2%",
          "inputPricePer1M": "$2",
          "outputPricePer1M": "$10",
          "pricePerSolvedTask": 2.35
        },
        {
          "name": "GPT-5.5",
          "maker": "OpenAI",
          "rank": 4,
          "open": false,
          "sweBenchVerified": "82.6%",
          "sweBenchPro": "58.6%",
          "inputPricePer1M": "$5",
          "outputPricePer1M": "$30",
          "pricePerSolvedTask": 6.05
        },
        {
          "name": "DeepSeek V4 Pro",
          "maker": "DeepSeek",
          "rank": 5,
          "open": true,
          "sweBenchVerified": "80.6%",
          "sweBenchPro": "55.4%",
          "inputPricePer1M": "$0.435",
          "outputPricePer1M": "$0.87",
          "pricePerSolvedTask": 0.54
        },
        {
          "name": "Gemini 3.1 Pro",
          "maker": "Google DeepMind",
          "rank": 6,
          "open": false,
          "sweBenchVerified": "80.6%",
          "sweBenchPro": "54.2%",
          "inputPricePer1M": "$2",
          "outputPricePer1M": "$12",
          "pricePerSolvedTask": 2.48
        },
        {
          "name": "MiniMax M3",
          "maker": "MiniMax",
          "rank": 7,
          "open": true,
          "sweBenchVerified": "80.5%",
          "sweBenchPro": "59.0%",
          "inputPricePer1M": "$0.60",
          "outputPricePer1M": null,
          "pricePerSolvedTask": 0.75
        },
        {
          "name": "Qwen3.7 Max",
          "maker": "Alibaba",
          "rank": 8,
          "open": false,
          "sweBenchVerified": "80.4%",
          "sweBenchPro": "60.6%",
          "inputPricePer1M": "$1.25",
          "outputPricePer1M": "$3.75",
          "pricePerSolvedTask": 1.55
        },
        {
          "name": "Kimi K2.6",
          "maker": "Moonshot AI",
          "rank": 9,
          "open": true,
          "sweBenchVerified": "80.2%",
          "sweBenchPro": "58.6%",
          "inputPricePer1M": "$0.95",
          "outputPricePer1M": "$4",
          "pricePerSolvedTask": 1.18
        },
        {
          "name": "Gemini 3.5 Flash",
          "maker": "Google DeepMind",
          "rank": 10,
          "open": false,
          "sweBenchVerified": "78.8%",
          "sweBenchPro": null,
          "inputPricePer1M": "$1.50",
          "outputPricePer1M": "$9",
          "pricePerSolvedTask": 1.9
        }
      ]
    },
    {
      "date": "2026-07-17",
      "commit": "829e59f20",
      "updated": "2026-07-17",
      "rankedCount": 13,
      "leader": "GPT-5.6 Sol",
      "models": [
        {
          "name": "GPT-5.6 Sol",
          "maker": "OpenAI",
          "rank": 1,
          "open": false,
          "sweBenchVerified": "96.2%",
          "sweBenchPro": null,
          "inputPricePer1M": "$5",
          "outputPricePer1M": "$30",
          "pricePerSolvedTask": 5.2
        },
        {
          "name": "Claude Fable 5",
          "maker": "Anthropic",
          "rank": 2,
          "open": false,
          "sweBenchVerified": "95.0%",
          "sweBenchPro": "80.3%",
          "inputPricePer1M": "$10",
          "outputPricePer1M": "$50",
          "pricePerSolvedTask": 10.53
        },
        {
          "name": "Claude Opus 4.8",
          "maker": "Anthropic",
          "rank": 3,
          "open": false,
          "sweBenchVerified": "88.6%",
          "sweBenchPro": "69.2%",
          "inputPricePer1M": "$5",
          "outputPricePer1M": "$25",
          "pricePerSolvedTask": 5.64
        },
        {
          "name": "Grok 4.5",
          "maker": "SpaceXAI (xAI)",
          "rank": 4,
          "open": false,
          "sweBenchVerified": "86.6%",
          "sweBenchPro": "64.7%",
          "inputPricePer1M": "$2",
          "outputPricePer1M": "$6",
          "pricePerSolvedTask": 2.31
        },
        {
          "name": "Claude Sonnet 5",
          "maker": "Anthropic",
          "rank": 5,
          "open": false,
          "sweBenchVerified": "85.2%",
          "sweBenchPro": "63.2%",
          "inputPricePer1M": "$2",
          "outputPricePer1M": "$10",
          "pricePerSolvedTask": 2.35
        },
        {
          "name": "GPT-5.5",
          "maker": "OpenAI",
          "rank": 6,
          "open": false,
          "sweBenchVerified": "82.6%",
          "sweBenchPro": "58.6%",
          "inputPricePer1M": "$5",
          "outputPricePer1M": "$30",
          "pricePerSolvedTask": 6.05
        },
        {
          "name": "Muse Spark 1.1",
          "maker": "Meta",
          "rank": 7,
          "open": false,
          "sweBenchVerified": "82.0%",
          "sweBenchPro": null,
          "inputPricePer1M": "$1.25",
          "outputPricePer1M": "$4.25",
          "pricePerSolvedTask": 1.52
        },
        {
          "name": "DeepSeek V4 Pro",
          "maker": "DeepSeek",
          "rank": 8,
          "open": true,
          "sweBenchVerified": "80.6%",
          "sweBenchPro": "55.4%",
          "inputPricePer1M": "$0.435",
          "outputPricePer1M": "$0.87",
          "pricePerSolvedTask": 0.54
        },
        {
          "name": "Gemini 3.1 Pro",
          "maker": "Google DeepMind",
          "rank": 9,
          "open": false,
          "sweBenchVerified": "80.6%",
          "sweBenchPro": "54.2%",
          "inputPricePer1M": "$2",
          "outputPricePer1M": "$12",
          "pricePerSolvedTask": 2.48
        },
        {
          "name": "MiniMax M3",
          "maker": "MiniMax",
          "rank": 10,
          "open": true,
          "sweBenchVerified": "80.5%",
          "sweBenchPro": "59.0%",
          "inputPricePer1M": "$0.60",
          "outputPricePer1M": null,
          "pricePerSolvedTask": 0.75
        },
        {
          "name": "Qwen3.7 Max",
          "maker": "Alibaba",
          "rank": 11,
          "open": false,
          "sweBenchVerified": "80.4%",
          "sweBenchPro": "60.6%",
          "inputPricePer1M": "$1.25",
          "outputPricePer1M": "$3.75",
          "pricePerSolvedTask": 1.55
        },
        {
          "name": "Kimi K2.6",
          "maker": "Moonshot AI",
          "rank": 12,
          "open": true,
          "sweBenchVerified": "80.2%",
          "sweBenchPro": "58.6%",
          "inputPricePer1M": "$0.95",
          "outputPricePer1M": "$4",
          "pricePerSolvedTask": 1.18
        },
        {
          "name": "Gemini 3.5 Flash",
          "maker": "Google DeepMind",
          "rank": 13,
          "open": false,
          "sweBenchVerified": "78.8%",
          "sweBenchPro": null,
          "inputPricePer1M": "$1.50",
          "outputPricePer1M": "$9",
          "pricePerSolvedTask": 1.9
        }
      ]
    }
  ]
}