{
  "schemaVersion": "1.0",
  "release": "local-v38",
  "protocol": "0.38.0",
  "generatedAt": "2026-10-25T19:19:41Z",
  "caseId": "ai-workforce-transformation",
  "coverage": {
    "decks": 289,
    "flagged_decks": 33,
    "scheduled_pairs": 987
  },
  "humanValidation": {
    "comparable_pairs": 49,
    "old_absolute_agreement_rate": 0.845,
    "panel_agreement_rate": 0.918
  },
  "scoreMeaning": "estimated chance of beating a typical rated configuration, on a 0-100 scale",
  "rows": [
    {
      "rank": 1,
      "configuration": "Claude Opus 5.5 / XHigh",
      "modelId": "claude-opus-5-5",
      "effort": "xhigh",
      "winScore": 97.9,
      "interval95": [
        95.7,
        99.5
      ],
      "elo": 2167.9,
      "decks": 7,
      "comparisons": 100,
      "costUsd": 6.543357,
      "costSamples": 7,
      "costSources": {
        "provider-reported": 7
      },
      "generationSeconds": 1341.641301,
      "generationSamples": 7
    },
    {
      "rank": 2,
      "configuration": "Claude Opus 5.5 / Max",
      "modelId": "claude-opus-5-5",
      "effort": "max",
      "winScore": 97.4,
      "interval95": [
        95.6,
        99
      ],
      "elo": 2130.4,
      "decks": 7,
      "comparisons": 126,
      "costUsd": 12.864245,
      "costSamples": 7,
      "costSources": {
        "provider-reported": 7
      },
      "generationSeconds": 2399.953174,
      "generationSamples": 7
    },
    {
      "rank": 3,
      "configuration": "Claude Sonnet 5.5 / Max",
      "modelId": "claude-sonnet-5-5",
      "effort": "max",
      "winScore": 96.9,
      "interval95": [
        94.5,
        98.9
      ],
      "elo": 2098.8,
      "decks": 5,
      "comparisons": 82,
      "costUsd": 21.394856,
      "costSamples": 5,
      "costSources": {
        "provider-reported": 5
      },
      "generationSeconds": 4375.880215,
      "generationSamples": 5
    },
    {
      "rank": 4,
      "configuration": "GPT-6 Astra / XHigh",
      "modelId": "gpt-6-astra",
      "effort": "xhigh",
      "winScore": 94.3,
      "interval95": [
        91.2,
        97.6
      ],
      "elo": 1989,
      "decks": 8,
      "comparisons": 100,
      "costUsd": 3.336314,
      "costSamples": 8,
      "costSources": {
        "api-equivalent-estimate": 8
      },
      "generationSeconds": 647.218377,
      "generationSamples": 8
    },
    {
      "rank": 5,
      "configuration": "GPT-6 Astra / Ultra",
      "modelId": "gpt-6-astra",
      "effort": "ultra",
      "winScore": 94.2,
      "interval95": [
        78.7,
        98.8
      ],
      "elo": 1984.7,
      "decks": 1,
      "comparisons": 8,
      "costUsd": 2.580394,
      "costSamples": 1,
      "costSources": {
        "api-equivalent-estimate": 1
      },
      "generationSeconds": 489.862821,
      "generationSamples": 1
    },
    {
      "rank": 6,
      "configuration": "GPT-5.6 Sol / Max",
      "modelId": "gpt-5.6-sol",
      "effort": "max",
      "winScore": 94,
      "interval95": [
        88.6,
        97.8
      ],
      "elo": 1977.5,
      "decks": 8,
      "comparisons": 80,
      "costUsd": 3.047264,
      "costSamples": 8,
      "costSources": {
        "api-equivalent-estimate": 8
      },
      "generationSeconds": 1064.930645,
      "generationSamples": 8
    },
    {
      "rank": 7,
      "configuration": "Claude Opus 5.5 / High",
      "modelId": "claude-opus-5-5",
      "effort": "high",
      "winScore": 93.6,
      "interval95": [
        89.8,
        97.1
      ],
      "elo": 1967.1,
      "decks": 6,
      "comparisons": 72,
      "costUsd": 2.872779,
      "costSamples": 6,
      "costSources": {
        "provider-reported": 6
      },
      "generationSeconds": 631.695912,
      "generationSamples": 6
    },
    {
      "rank": 8,
      "configuration": "GPT-5.6 Sol / Ultra",
      "modelId": "gpt-5.6-sol",
      "effort": "ultra",
      "winScore": 93.3,
      "interval95": [
        68.4,
        99.3
      ],
      "elo": 1958.9,
      "decks": 1,
      "comparisons": 8,
      "costUsd": 3.376012,
      "costSamples": 1,
      "costSources": {
        "api-equivalent-estimate": 1
      },
      "generationSeconds": 951.773594,
      "generationSamples": 1
    },
    {
      "rank": 9,
      "configuration": "GPT-6 Astra / Max",
      "modelId": "gpt-6-astra",
      "effort": "max",
      "winScore": 92,
      "interval95": [
        86.8,
        96.9
      ],
      "elo": 1924.4,
      "decks": 6,
      "comparisons": 74,
      "costUsd": 4.492284,
      "costSamples": 6,
      "costSources": {
        "api-equivalent-estimate": 6
      },
      "generationSeconds": 1040.953669,
      "generationSamples": 6
    },
    {
      "rank": 10,
      "configuration": "Claude Sonnet 5.5 / XHigh",
      "modelId": "claude-sonnet-5-5",
      "effort": "xhigh",
      "winScore": 90.6,
      "interval95": [
        83.5,
        95.5
      ],
      "elo": 1893.4,
      "decks": 4,
      "comparisons": 58,
      "costUsd": 7.039239,
      "costSamples": 4,
      "costSources": {
        "provider-reported": 4
      },
      "generationSeconds": 1517.65777,
      "generationSamples": 4
    },
    {
      "rank": 11,
      "configuration": "DeepSeek V4.1 Flash 1017 / Max",
      "modelId": "deepseek/deepseek-v4.1-flash-1017",
      "effort": "max",
      "winScore": 89.8,
      "interval95": [
        77.7,
        97.6
      ],
      "elo": 1877.7,
      "decks": 3,
      "comparisons": 26,
      "costUsd": 0.249929,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 1614.585692,
      "generationSamples": 3
    },
    {
      "rank": 12,
      "configuration": "Kimi K3.1 Pro / High",
      "modelId": "moonshotai/kimi-k3.1-pro",
      "effort": "high",
      "winScore": 89.6,
      "interval95": [
        80.5,
        97.1
      ],
      "elo": 1873.8,
      "decks": 3,
      "comparisons": 30,
      "costUsd": 1.1122,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 795.940808,
      "generationSamples": 3
    },
    {
      "rank": 13,
      "configuration": "Kimi K3.1 Pro / Max",
      "modelId": "moonshotai/kimi-k3.1-pro",
      "effort": "max",
      "winScore": 87.7,
      "interval95": [
        71.4,
        95.2
      ],
      "elo": 1840.6,
      "decks": 3,
      "comparisons": 34,
      "costUsd": 1.786245,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 1345.517627,
      "generationSamples": 3
    },
    {
      "rank": 14,
      "configuration": "Claude Opus 5.5 / Medium",
      "modelId": "claude-opus-5-5",
      "effort": "medium",
      "winScore": 80.1,
      "interval95": [
        66.7,
        90.3
      ],
      "elo": 1742.4,
      "decks": 4,
      "comparisons": 70,
      "costUsd": 1.731365,
      "costSamples": 4,
      "costSources": {
        "provider-reported": 4
      },
      "generationSeconds": 478.094867,
      "generationSamples": 4
    },
    {
      "rank": 15,
      "configuration": "GPT-6 Sol / Max",
      "modelId": "gpt-6-sol",
      "effort": "max",
      "winScore": 80.1,
      "interval95": [
        65.1,
        91.7
      ],
      "elo": 1741.4,
      "decks": 5,
      "comparisons": 54,
      "costUsd": 0.896665,
      "costSamples": 5,
      "costSources": {
        "api-equivalent-estimate": 5
      },
      "generationSeconds": 662.246974,
      "generationSamples": 5
    },
    {
      "rank": 16,
      "configuration": "Claude Sonnet 5.5 / High",
      "modelId": "claude-sonnet-5-5",
      "effort": "high",
      "winScore": 79.8,
      "interval95": [
        67,
        91.7
      ],
      "elo": 1738.9,
      "decks": 3,
      "comparisons": 60,
      "costUsd": 1.574966,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 324.638297,
      "generationSamples": 3
    },
    {
      "rank": 17,
      "configuration": "GPT-6.1 Sol / Max",
      "modelId": "gpt-6.1-sol",
      "effort": "max",
      "winScore": 79.3,
      "interval95": [
        70.7,
        87.4
      ],
      "elo": 1733.2,
      "decks": 5,
      "comparisons": 96,
      "costUsd": 0.891597,
      "costSamples": 5,
      "costSources": {
        "api-equivalent-estimate": 5
      },
      "generationSeconds": 1363.291837,
      "generationSamples": 5
    },
    {
      "rank": 18,
      "configuration": "GPT-6 Astra / High",
      "modelId": "gpt-6-astra",
      "effort": "high",
      "winScore": 78.6,
      "interval95": [
        65.7,
        88
      ],
      "elo": 1726.1,
      "decks": 5,
      "comparisons": 52,
      "costUsd": 2.357097,
      "costSamples": 5,
      "costSources": {
        "api-equivalent-estimate": 5
      },
      "generationSeconds": 392.932999,
      "generationSamples": 5
    },
    {
      "rank": 19,
      "configuration": "GPT-6.1 Sol / Ultra",
      "modelId": "gpt-6.1-sol",
      "effort": "ultra",
      "winScore": 78.3,
      "interval95": [
        42.5,
        95.3
      ],
      "elo": 1723.3,
      "decks": 3,
      "comparisons": 24,
      "costUsd": 0.626828,
      "costSamples": 3,
      "costSources": {
        "api-equivalent-estimate": 3
      },
      "generationSeconds": 697.624622,
      "generationSamples": 3
    },
    {
      "rank": 20,
      "configuration": "Claude Opus 5 / XHigh",
      "modelId": "claude-opus-5",
      "effort": "xhigh",
      "winScore": 76.6,
      "interval95": [
        55.6,
        90.5
      ],
      "elo": 1705.7,
      "decks": 2,
      "comparisons": 30,
      "costUsd": 10.387651,
      "costSamples": 2,
      "costSources": {
        "provider-reported": 2
      },
      "generationSeconds": 1669.426636,
      "generationSamples": 2
    },
    {
      "rank": 21,
      "configuration": "Claude Haiku 5.5 / High",
      "modelId": "claude-haiku-5-5",
      "effort": "high",
      "winScore": 75.4,
      "interval95": [
        55.7,
        87.5
      ],
      "elo": 1695,
      "decks": 5,
      "comparisons": 48,
      "costUsd": 2.357798,
      "costSamples": 5,
      "costSources": {
        "provider-reported": 5
      },
      "generationSeconds": 517.052301,
      "generationSamples": 5
    },
    {
      "rank": 22,
      "configuration": "Claude Opus 4.8 / Max",
      "modelId": "claude-opus-4-8",
      "effort": "max",
      "winScore": 75,
      "interval95": [
        60.8,
        85.8
      ],
      "elo": 1690.5,
      "decks": 2,
      "comparisons": 26,
      "costUsd": 11.272232,
      "costSamples": 2,
      "costSources": {
        "provider-reported": 2
      },
      "generationSeconds": 2055.012992,
      "generationSamples": 2
    },
    {
      "rank": 23,
      "configuration": "Kimi K3.1 Pro / Low",
      "modelId": "moonshotai/kimi-k3.1-pro",
      "effort": "low",
      "winScore": 73.8,
      "interval95": [
        48.1,
        92
      ],
      "elo": 1680.3,
      "decks": 3,
      "comparisons": 24,
      "costUsd": 1.3378,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 1227.409434,
      "generationSamples": 3
    },
    {
      "rank": 24,
      "configuration": "GPT-5.6 Sol / XHigh",
      "modelId": "gpt-5.6-sol",
      "effort": "xhigh",
      "winScore": 73.2,
      "interval95": [
        58.7,
        84.3
      ],
      "elo": 1674.8,
      "decks": 4,
      "comparisons": 60,
      "costUsd": 2.361614,
      "costSamples": 4,
      "costSources": {
        "api-equivalent-estimate": 4
      },
      "generationSeconds": 1438.592786,
      "generationSamples": 4
    },
    {
      "rank": 25,
      "configuration": "MiniMax M3.1 Flash / Max",
      "modelId": "minimax/minimax-m3.1-flash",
      "effort": "max",
      "winScore": 70.7,
      "interval95": [
        48.2,
        89.4
      ],
      "elo": 1653.3,
      "decks": 1,
      "comparisons": 8,
      "costUsd": 0.288957,
      "costSamples": 1,
      "costSources": {
        "provider-reported": 1
      },
      "generationSeconds": 2577.756281,
      "generationSamples": 1
    },
    {
      "rank": 26,
      "configuration": "GPT-6.1 Sol / Medium",
      "modelId": "gpt-6.1-sol",
      "effort": "medium",
      "winScore": 66.8,
      "interval95": [
        53.2,
        80.5
      ],
      "elo": 1621.5,
      "decks": 5,
      "comparisons": 72,
      "costUsd": 0.377323,
      "costSamples": 5,
      "costSources": {
        "api-equivalent-estimate": 5
      },
      "generationSeconds": 396.931875,
      "generationSamples": 5
    },
    {
      "rank": 27,
      "configuration": "Claude Opus 4.8 / XHigh",
      "modelId": "claude-opus-4-8",
      "effort": "xhigh",
      "winScore": 65.1,
      "interval95": [
        40.5,
        83.6
      ],
      "elo": 1608.5,
      "decks": 3,
      "comparisons": 24,
      "costUsd": 6.763447,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 1341.344113,
      "generationSamples": 3
    },
    {
      "rank": 28,
      "configuration": "Claude Haiku 5.5 / Medium",
      "modelId": "claude-haiku-5-5",
      "effort": "medium",
      "winScore": 64.3,
      "interval95": [
        52,
        75.4
      ],
      "elo": 1602,
      "decks": 6,
      "comparisons": 70,
      "costUsd": 2.389728,
      "costSamples": 6,
      "costSources": {
        "provider-reported": 6
      },
      "generationSeconds": 367.723323,
      "generationSamples": 6
    },
    {
      "rank": 29,
      "configuration": "Claude Haiku 5.5 / Max",
      "modelId": "claude-haiku-5-5",
      "effort": "max",
      "winScore": 61.4,
      "interval95": [
        42.5,
        77.2
      ],
      "elo": 1580.9,
      "decks": 3,
      "comparisons": 32,
      "costUsd": 20.685858,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 2781.759669,
      "generationSamples": 3
    },
    {
      "rank": 30,
      "configuration": "GPT-6 Astra / Low",
      "modelId": "gpt-6-astra",
      "effort": "low",
      "winScore": 59.7,
      "interval95": [
        43.3,
        75.1
      ],
      "elo": 1568.3,
      "decks": 5,
      "comparisons": 70,
      "costUsd": 1.467947,
      "costSamples": 5,
      "costSources": {
        "api-equivalent-estimate": 5
      },
      "generationSeconds": 274.434238,
      "generationSamples": 5
    },
    {
      "rank": 31,
      "configuration": "GPT-6 Sol / XHigh",
      "modelId": "gpt-6-sol",
      "effort": "xhigh",
      "winScore": 59.3,
      "interval95": [
        45.8,
        73.4
      ],
      "elo": 1565.5,
      "decks": 4,
      "comparisons": 80,
      "costUsd": 0.874798,
      "costSamples": 4,
      "costSources": {
        "api-equivalent-estimate": 4
      },
      "generationSeconds": 1278.255309,
      "generationSamples": 4
    },
    {
      "rank": 32,
      "configuration": "Claude Sonnet 5.5 / Medium",
      "modelId": "claude-sonnet-5-5",
      "effort": "medium",
      "winScore": 59.2,
      "interval95": [
        40.1,
        76.4
      ],
      "elo": 1564.4,
      "decks": 3,
      "comparisons": 44,
      "costUsd": 0.939375,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 196.522116,
      "generationSamples": 3
    },
    {
      "rank": 33,
      "configuration": "GPT-6.1 Sol / Low",
      "modelId": "gpt-6.1-sol",
      "effort": "low",
      "winScore": 58.7,
      "interval95": [
        43.5,
        71.6
      ],
      "elo": 1560.9,
      "decks": 7,
      "comparisons": 80,
      "costUsd": 0.21583,
      "costSamples": 7,
      "costSources": {
        "api-equivalent-estimate": 7
      },
      "generationSeconds": 271.093422,
      "generationSamples": 7
    },
    {
      "rank": 34,
      "configuration": "GPT-6.1 Sol / XHigh",
      "modelId": "gpt-6.1-sol",
      "effort": "xhigh",
      "winScore": 57.9,
      "interval95": [
        42.2,
        70.7
      ],
      "elo": 1555.6,
      "decks": 5,
      "comparisons": 92,
      "costUsd": 0.647418,
      "costSamples": 5,
      "costSources": {
        "api-equivalent-estimate": 5
      },
      "generationSeconds": 837.863612,
      "generationSamples": 5
    },
    {
      "rank": 35,
      "configuration": "Claude Fable 5.1 / Max",
      "modelId": "claude-fable-5-1",
      "effort": "max",
      "winScore": 54.8,
      "interval95": [
        40,
        68.8
      ],
      "elo": 1533.6,
      "decks": 4,
      "comparisons": 62,
      "costUsd": 7.342822,
      "costSamples": 4,
      "costSources": {
        "api-equivalent-estimate": 3,
        "provider-reported": 1
      },
      "generationSeconds": 725.2918,
      "generationSamples": 4
    },
    {
      "rank": 36,
      "configuration": "Claude Opus 5 / Max",
      "modelId": "claude-opus-5",
      "effort": "max",
      "winScore": 54.4,
      "interval95": [
        32.9,
        74.2
      ],
      "elo": 1530.9,
      "decks": 3,
      "comparisons": 28,
      "costUsd": 16.453612,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 2291.890414,
      "generationSamples": 3
    },
    {
      "rank": 37,
      "configuration": "GPT-5.6 Sol / High",
      "modelId": "gpt-5.6-sol",
      "effort": "high",
      "winScore": 52.9,
      "interval95": [
        34.7,
        66.5
      ],
      "elo": 1520.1,
      "decks": 5,
      "comparisons": 78,
      "costUsd": 1.553284,
      "costSamples": 5,
      "costSources": {
        "api-equivalent-estimate": 5
      },
      "generationSeconds": 1291.681617,
      "generationSamples": 5
    },
    {
      "rank": 38,
      "configuration": "GPT-6 Astra / Medium",
      "modelId": "gpt-6-astra",
      "effort": "medium",
      "winScore": 52.1,
      "interval95": [
        37.9,
        66
      ],
      "elo": 1514.5,
      "decks": 4,
      "comparisons": 82,
      "costUsd": 1.345792,
      "costSamples": 4,
      "costSources": {
        "api-equivalent-estimate": 4
      },
      "generationSeconds": 273.095581,
      "generationSamples": 4
    },
    {
      "rank": 39,
      "configuration": "GPT-6 Sol / High",
      "modelId": "gpt-6-sol",
      "effort": "high",
      "winScore": 51.7,
      "interval95": [
        29.6,
        71.4
      ],
      "elo": 1511.6,
      "decks": 5,
      "comparisons": 48,
      "costUsd": 0.577283,
      "costSamples": 5,
      "costSources": {
        "api-equivalent-estimate": 5
      },
      "generationSeconds": 333.756326,
      "generationSamples": 5
    },
    {
      "rank": 40,
      "configuration": "GPT-6.1 Sol / High",
      "modelId": "gpt-6.1-sol",
      "effort": "high",
      "winScore": 50.9,
      "interval95": [
        38.1,
        65.4
      ],
      "elo": 1506.6,
      "decks": 5,
      "comparisons": 86,
      "costUsd": 0.557545,
      "costSamples": 5,
      "costSources": {
        "api-equivalent-estimate": 5
      },
      "generationSeconds": 842.483135,
      "generationSamples": 5
    },
    {
      "rank": 41,
      "configuration": "GLM 5.3 / Max",
      "modelId": "z-ai/glm-5.3",
      "effort": "max",
      "winScore": 47.2,
      "interval95": [
        14,
        83
      ],
      "elo": 1480.5,
      "decks": 1,
      "comparisons": 8,
      "costUsd": 0.624128,
      "costSamples": 1,
      "costSources": {
        "provider-reported": 1
      },
      "generationSeconds": 921.277821,
      "generationSamples": 1
    },
    {
      "rank": 42,
      "configuration": "Claude Haiku 5.5 / XHigh",
      "modelId": "claude-haiku-5-5",
      "effort": "xhigh",
      "winScore": 44.8,
      "interval95": [
        21.6,
        67.6
      ],
      "elo": 1463.4,
      "decks": 3,
      "comparisons": 36,
      "costUsd": 6.737928,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 911.915065,
      "generationSamples": 3
    },
    {
      "rank": 43,
      "configuration": "GPT-6 Luna / Max",
      "modelId": "gpt-6-luna",
      "effort": "max",
      "winScore": 44.5,
      "interval95": [
        35,
        54
      ],
      "elo": 1461.9,
      "decks": 10,
      "comparisons": 154,
      "costUsd": 0.083696,
      "costSamples": 10,
      "costSources": {
        "api-equivalent-estimate": 10
      },
      "generationSeconds": 1522.289673,
      "generationSamples": 10
    },
    {
      "rank": 44,
      "configuration": "MiMo-V2.6-Flash / Max",
      "modelId": "mimo-v2.6-flash",
      "effort": "max",
      "winScore": 43.9,
      "interval95": [
        26,
        63.7
      ],
      "elo": 1457.7,
      "decks": 2,
      "comparisons": 34,
      "costUsd": 0.158183,
      "costSamples": 2,
      "costSources": {
        "provider-reported": 2
      },
      "generationSeconds": 3572.316956,
      "generationSamples": 2
    },
    {
      "rank": 45,
      "configuration": "DeepSeek V4.1 Flash / Max",
      "modelId": "deepseek/deepseek-v4.1-flash",
      "effort": "max",
      "winScore": 42.5,
      "interval95": [
        30.9,
        56.5
      ],
      "elo": 1447.1,
      "decks": 6,
      "comparisons": 108,
      "costUsd": 0.133543,
      "costSamples": 6,
      "costSources": {
        "provider-reported": 6
      },
      "generationSeconds": 886.45917,
      "generationSamples": 6
    },
    {
      "rank": 46,
      "configuration": "Grok 4.6 / Max",
      "modelId": "x-ai/grok-4.6",
      "effort": "max",
      "winScore": 42.2,
      "interval95": [
        28.7,
        58.4
      ],
      "elo": 1445.4,
      "decks": 5,
      "comparisons": 88,
      "costUsd": 1.688598,
      "costSamples": 5,
      "costSources": {
        "provider-reported": 5
      },
      "generationSeconds": 1345.068709,
      "generationSamples": 5
    },
    {
      "rank": 47,
      "configuration": "GPT-6 Sol / Ultra",
      "modelId": "gpt-6-sol",
      "effort": "ultra",
      "winScore": 40.2,
      "interval95": [
        11.5,
        79.6
      ],
      "elo": 1431.4,
      "decks": 1,
      "comparisons": 8,
      "costUsd": 1.624225,
      "costSamples": 1,
      "costSources": {
        "api-equivalent-estimate": 1
      },
      "generationSeconds": 630.102646,
      "generationSamples": 1
    },
    {
      "rank": 48,
      "configuration": "GLM 5.3 Flash / Max",
      "modelId": "z-ai/glm-5.3-flash",
      "effort": "max",
      "winScore": 37,
      "interval95": [
        20.1,
        54.9
      ],
      "elo": 1407.9,
      "decks": 3,
      "comparisons": 74,
      "costUsd": 0.23913,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 1661.638611,
      "generationSamples": 3
    },
    {
      "rank": 49,
      "configuration": "GPT-6 Sol / Medium",
      "modelId": "gpt-6-sol",
      "effort": "medium",
      "winScore": 35.4,
      "interval95": [
        23.2,
        48.3
      ],
      "elo": 1395.7,
      "decks": 5,
      "comparisons": 82,
      "costUsd": 0.349884,
      "costSamples": 5,
      "costSources": {
        "api-equivalent-estimate": 5
      },
      "generationSeconds": 229.140128,
      "generationSamples": 5
    },
    {
      "rank": 50,
      "configuration": "GPT-5.6 Terra / Max",
      "modelId": "gpt-5.6-terra",
      "effort": "max",
      "winScore": 34.5,
      "interval95": [
        19.8,
        52.9
      ],
      "elo": 1388.9,
      "decks": 5,
      "comparisons": 64,
      "costUsd": 1.64145,
      "costSamples": 5,
      "costSources": {
        "api-equivalent-estimate": 5
      },
      "generationSeconds": 927.283994,
      "generationSamples": 5
    },
    {
      "rank": 51,
      "configuration": "Qwen 3.8 Flash / Max",
      "modelId": "qwen/qwen3.8-flash",
      "effort": "max",
      "winScore": 31.7,
      "interval95": [
        15.4,
        52.8
      ],
      "elo": 1367,
      "decks": 3,
      "comparisons": 44,
      "costUsd": 0.189307,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 1807.890884,
      "generationSamples": 3
    },
    {
      "rank": 52,
      "configuration": "Kimi K3 / Max",
      "modelId": "moonshotai/kimi-k3",
      "effort": "max",
      "winScore": 27.5,
      "interval95": [
        5.4,
        62.9
      ],
      "elo": 1331.3,
      "decks": 1,
      "comparisons": 12,
      "costUsd": 1.733997,
      "costSamples": 1,
      "costSources": {
        "provider-reported": 1
      },
      "generationSeconds": 818.430901,
      "generationSamples": 1
    },
    {
      "rank": 53,
      "configuration": "GPT-6 Luna / XHigh",
      "modelId": "gpt-6-luna",
      "effort": "xhigh",
      "winScore": 25.4,
      "interval95": [
        14.2,
        37.8
      ],
      "elo": 1312.4,
      "decks": 4,
      "comparisons": 74,
      "costUsd": 0.077345,
      "costSamples": 4,
      "costSources": {
        "api-equivalent-estimate": 4
      },
      "generationSeconds": 1130.360766,
      "generationSamples": 4
    },
    {
      "rank": 54,
      "configuration": "Claude Haiku 5.5 / Low",
      "modelId": "claude-haiku-5-5",
      "effort": "low",
      "winScore": 20.5,
      "interval95": [
        6.1,
        40.9
      ],
      "elo": 1264.3,
      "decks": 3,
      "comparisons": 30,
      "costUsd": 2.123343,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 270.182983,
      "generationSamples": 3
    },
    {
      "rank": 55,
      "configuration": "GPT-5.5 / XHigh",
      "modelId": "gpt-5.5",
      "effort": "xhigh",
      "winScore": 18.3,
      "interval95": [
        8.4,
        35.1
      ],
      "elo": 1240.4,
      "decks": 3,
      "comparisons": 42,
      "costUsd": 2.560894,
      "costSamples": 3,
      "costSources": {
        "api-equivalent-estimate": 3
      },
      "generationSeconds": 689.328012,
      "generationSamples": 3
    },
    {
      "rank": 56,
      "configuration": "GPT-6 Luna / High",
      "modelId": "gpt-6-luna",
      "effort": "high",
      "winScore": 14,
      "interval95": [
        4.9,
        23.7
      ],
      "elo": 1184.6,
      "decks": 4,
      "comparisons": 38,
      "costUsd": 0.034769,
      "costSamples": 4,
      "costSources": {
        "api-equivalent-estimate": 4
      },
      "generationSeconds": 311.310442,
      "generationSamples": 4
    },
    {
      "rank": 57,
      "configuration": "GPT-6 Sol / Low",
      "modelId": "gpt-6-sol",
      "effort": "low",
      "winScore": 13.9,
      "interval95": [
        4.2,
        27.2
      ],
      "elo": 1183.8,
      "decks": 6,
      "comparisons": 70,
      "costUsd": 0.160489,
      "costSamples": 6,
      "costSources": {
        "api-equivalent-estimate": 6
      },
      "generationSeconds": 157.989206,
      "generationSamples": 6
    },
    {
      "rank": 58,
      "configuration": "Gemini 3.8 Flash / Max",
      "modelId": "google/gemini-3.8-flash",
      "effort": "max",
      "winScore": 13.8,
      "interval95": [
        5.4,
        25.9
      ],
      "elo": 1182.2,
      "decks": 4,
      "comparisons": 74,
      "costUsd": 2.020343,
      "costSamples": 4,
      "costSources": {
        "provider-reported": 4
      },
      "generationSeconds": 1072.113495,
      "generationSamples": 4
    },
    {
      "rank": 59,
      "configuration": "Muse Spark 1.3 / Max",
      "modelId": "meta/muse-spark-1.3",
      "effort": "max",
      "winScore": 13.8,
      "interval95": [
        6.5,
        21
      ],
      "elo": 1181.8,
      "decks": 6,
      "comparisons": 80,
      "costUsd": 0.721345,
      "costSamples": 6,
      "costSources": {
        "provider-reported": 6
      },
      "generationSeconds": 429.069272,
      "generationSamples": 6
    },
    {
      "rank": 60,
      "configuration": "GPT-6 Luna / Low",
      "modelId": "gpt-6-luna",
      "effort": "low",
      "winScore": 6.9,
      "interval95": [
        1.6,
        18.1
      ],
      "elo": 1047.5,
      "decks": 3,
      "comparisons": 34,
      "costUsd": 0.014933,
      "costSamples": 3,
      "costSources": {
        "api-equivalent-estimate": 3
      },
      "generationSeconds": 153.680157,
      "generationSamples": 3
    },
    {
      "rank": 61,
      "configuration": "Claude Opus 5.5 / Low",
      "modelId": "claude-opus-5-5",
      "effort": "low",
      "winScore": 6.7,
      "interval95": [
        2.2,
        11.8
      ],
      "elo": 1042.8,
      "decks": 3,
      "comparisons": 78,
      "costUsd": 0.4939,
      "costSamples": 3,
      "costSources": {
        "provider-reported": 3
      },
      "generationSeconds": 86.987104,
      "generationSamples": 3
    },
    {
      "rank": 62,
      "configuration": "GPT-6 Luna / Medium",
      "modelId": "gpt-6-luna",
      "effort": "medium",
      "winScore": 5.6,
      "interval95": [
        1,
        17.3
      ],
      "elo": 1009.7,
      "decks": 4,
      "comparisons": 42,
      "costUsd": 0.015936,
      "costSamples": 4,
      "costSources": {
        "api-equivalent-estimate": 4
      },
      "generationSeconds": 109.661947,
      "generationSamples": 4
    },
    {
      "rank": 63,
      "configuration": "GPT-5.6 Luna / XHigh",
      "modelId": "gpt-5.6-luna",
      "effort": "xhigh",
      "winScore": 4.4,
      "interval95": [
        1,
        11.2
      ],
      "elo": 966.9,
      "decks": 4,
      "comparisons": 54,
      "costUsd": 0.110245,
      "costSamples": 4,
      "costSources": {
        "api-equivalent-estimate": 4
      },
      "generationSeconds": 689.692378,
      "generationSamples": 4
    },
    {
      "rank": 64,
      "configuration": "DeepSeek V4 Pro 0813 / Max",
      "modelId": "deepseek/deepseek-v4-pro-0813",
      "effort": "max",
      "winScore": 4.3,
      "interval95": [
        1.7,
        14.9
      ],
      "elo": 962.5,
      "decks": 1,
      "comparisons": 12,
      "costUsd": 0.457819,
      "costSamples": 1,
      "costSources": {
        "provider-reported": 1
      },
      "generationSeconds": 1634.562061,
      "generationSamples": 1
    },
    {
      "rank": 65,
      "configuration": "Muse Wonder / Max",
      "modelId": "meta/muse-wonder",
      "effort": "max",
      "winScore": 1.7,
      "interval95": [
        0.3,
        25.7
      ],
      "elo": 793.6,
      "decks": 1,
      "comparisons": 10,
      "costUsd": 3.032301,
      "costSamples": 1,
      "costSources": {
        "provider-reported": 1
      },
      "generationSeconds": 682.077744,
      "generationSamples": 1
    },
    {
      "rank": 66,
      "configuration": "GPT-5.5 / High",
      "modelId": "gpt-5.5",
      "effort": "high",
      "winScore": 0.9,
      "interval95": [
        0.5,
        1.4
      ],
      "elo": 680,
      "decks": 3,
      "comparisons": 48,
      "costUsd": 1.756934,
      "costSamples": 3,
      "costSources": {
        "api-equivalent-estimate": 3
      },
      "generationSeconds": 448.182934,
      "generationSamples": 3
    }
  ],
  "summary": "https://www.andrewmcgehee.com/benchmarks/powerpoint/",
  "methodology": "https://www.andrewmcgehee.com/benchmarks/powerpoint/methodology/",
  "fullComparison": "https://www.andrewmcgehee.com/projects/powerpoint-benchmark/",
  "reuseTerms": "https://www.andrewmcgehee.com/legal/",
  "limitations": "One corporate-design task. Relative scores depend on the rated pool. Intervals and sample counts matter. Cost and time have their own sample counts. Retired models omitted from public listings remain in the rating pool. This is a current snapshot, not an archive or a complete reproduction pack."
}