{
  "trials": 108,
  "plannedTrials": 108,
  "complete": true,
  "totalCost": 2.1678536819999996,
  "byModel": {
    "openai/gpt-5.6-luna": {
      "plain": {
        "trials": 18,
        "successes": 18,
        "naturalCompletions": 18,
        "terminations": {
          "finish": 12,
          "text": 6
        },
        "totalCost": 0.024193026000000003,
        "meanCost": 0.001344057,
        "costPerSuccess": 0.001344057,
        "meanNoCacheCost": 0.003750049777777777,
        "meanSeconds": 16.952917868055515,
        "medianSeconds": 15.656218916499988,
        "p95Seconds": 28.39395384824967,
        "mainCalls": 112,
        "browserCalls": 94,
        "jevCalls": 101,
        "inputTokens": 278263,
        "cachedInputTokens": 248714,
        "cacheWriteTokens": 29213,
        "outputTokens": 5177,
        "reasoningTokens": 906,
        "jevInputTokens": 134188,
        "mainSeconds": 232.78942758299922,
        "browserSeconds": 72.35879016400064,
        "mainCost": 0.01855713,
        "jevCost": 0.0056358960000000005,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 5.421010862427522e-20
      },
      "playwright": {
        "trials": 18,
        "successes": 18,
        "naturalCompletions": 18,
        "terminations": {
          "finish": 9,
          "text": 9
        },
        "totalCost": 0.022710309999999994,
        "meanCost": 0.0012616838888888885,
        "costPerSuccess": 0.0012616838888888885,
        "meanNoCacheCost": 0.0041852666666666654,
        "meanSeconds": 35.366335238388935,
        "medianSeconds": 32.155958583499945,
        "p95Seconds": 69.25720141080028,
        "mainCalls": 137,
        "browserCalls": 123,
        "jevCalls": 0,
        "inputTokens": 345714,
        "cachedInputTokens": 303868,
        "cacheWriteTokens": 41435,
        "outputTokens": 5160,
        "reasoningTokens": 821,
        "jevInputTokens": 0,
        "mainSeconds": 510.8553106279985,
        "browserSeconds": 125.73300874500048,
        "mainCost": 0.022710309999999994,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.168404344971009e-19
      },
      "comparison": {
        "costRatio": 1.065288232525228,
        "costRatio95": [
          0.9667475996549272,
          1.1626448769783335
        ],
        "timeRatio": 0.4793518399286593,
        "timeRatio95": [
          0.36807094754300174,
          0.6223587345260233
        ]
      }
    },
    "openai/gpt-5.6-terra": {
      "plain": {
        "trials": 18,
        "successes": 18,
        "naturalCompletions": 18,
        "terminations": {
          "finish": 18
        },
        "totalCost": 0.18331519400000001,
        "meanCost": 0.010184177444444446,
        "costPerSuccess": 0.010184177444444446,
        "meanNoCacheCost": 0.034290877444444444,
        "meanSeconds": 17.03366856705557,
        "medianSeconds": 16.434109979000045,
        "p95Seconds": 24.016077795650066,
        "mainCalls": 113,
        "browserCalls": 95,
        "jevCalls": 101,
        "inputTokens": 276895,
        "cachedInputTokens": 248782,
        "cacheWriteTokens": 27774,
        "outputTokens": 4819,
        "reasoningTokens": 637,
        "jevInputTokens": 133757,
        "mainSeconds": 232.8610701239987,
        "browserSeconds": 73.73995220700048,
        "mainCost": 0.1776974,
        "jevCost": 0.0056177940000000015,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 8.673617379884035e-19
      },
      "playwright": {
        "trials": 18,
        "successes": 17,
        "naturalCompletions": 17,
        "terminations": {
          "text": 2,
          "finish": 16
        },
        "totalCost": 0.24072589999999996,
        "meanCost": 0.013373661111111108,
        "costPerSuccess": 0.014160347058823528,
        "meanNoCacheCost": 0.04665788888888889,
        "meanSeconds": 22.85069523594442,
        "medianSeconds": 22.146628875000026,
        "p95Seconds": 35.80702760449987,
        "mainCalls": 143,
        "browserCalls": 127,
        "jevCalls": 0,
        "inputTokens": 389183,
        "cachedInputTokens": 344997,
        "cacheWriteTokens": 43757,
        "outputTokens": 5123,
        "reasoningTokens": 863,
        "jevInputTokens": 0,
        "mainSeconds": 279.46122966199886,
        "browserSeconds": 131.8452515840024,
        "mainCost": 0.24072589999999996,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 1.734723475976807e-18
      },
      "comparison": {
        "costRatio": 0.7615100577046343,
        "costRatio95": [
          0.6635927849383548,
          0.9067622385455407
        ],
        "timeRatio": 0.7454332741815841,
        "timeRatio95": [
          0.6619233274957169,
          0.94343589829332
        ]
      }
    },
    "openai/gpt-6-astra": {
      "plain": {
        "trials": 18,
        "successes": 18,
        "naturalCompletions": 18,
        "terminations": {
          "finish": 18
        },
        "totalCost": 0.802392252,
        "meanCost": 0.04457734733333333,
        "costPerSuccess": 0.04457734733333333,
        "meanNoCacheCost": 0.1585875417777778,
        "meanSeconds": 20.401571705999977,
        "medianSeconds": 19.24601254200001,
        "p95Seconds": 27.847826306200012,
        "mainCalls": 108,
        "browserCalls": 90,
        "jevCalls": 70,
        "inputTokens": 268312,
        "cachedInputTokens": 236709,
        "cacheWriteTokens": 31279,
        "outputTokens": 3368,
        "reasoningTokens": 0,
        "jevInputTokens": 72756,
        "mainSeconds": 304.20344312899965,
        "browserSeconds": 63.01989424799923,
        "mainCost": 0.7993365000000002,
        "jevCost": 0.0030557520000000006,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 3.469446951953614e-18
      },
      "playwright": {
        "trials": 18,
        "successes": 18,
        "naturalCompletions": 18,
        "terminations": {
          "finish": 18
        },
        "totalCost": 0.894517,
        "meanCost": 0.049695388888888886,
        "costPerSuccess": 0.049695388888888886,
        "meanNoCacheCost": 0.17827388888888887,
        "meanSeconds": 41.76455107872223,
        "medianSeconds": 34.127917332999985,
        "p95Seconds": 79.63807185179999,
        "mainCalls": 132,
        "browserCalls": 114,
        "jevCalls": 0,
        "inputTokens": 302748,
        "cachedInputTokens": 266982,
        "cacheWriteTokens": 35370,
        "outputTokens": 3629,
        "reasoningTokens": 0,
        "jevInputTokens": 0,
        "mainSeconds": 626.7908222419992,
        "browserSeconds": 124.96619779099768,
        "mainCost": 0.894517,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 1.734723475976807e-18
      },
      "comparison": {
        "costRatio": 0.8970117415320222,
        "costRatio95": [
          0.7667083524792284,
          1.057576591363709
        ],
        "timeRatio": 0.488490147243411,
        "timeRatio95": [
          0.380147942196216,
          0.6470593800157557
        ]
      }
    }
  },
  "byTask": {
    "openai/gpt-5.6-luna/contact": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.002622524,
        "meanCost": 0.0008741746666666668,
        "costPerSuccess": 0.0008741746666666668,
        "meanNoCacheCost": 0.0026417079999999996,
        "meanSeconds": 12.078752277999882,
        "medianSeconds": 11.075029458999634,
        "p95Seconds": 14.092226996199972,
        "mainCalls": 16,
        "browserCalls": 13,
        "jevCalls": 14,
        "inputTokens": 33537,
        "cachedInputTokens": 30335,
        "cacheWriteTokens": 3154,
        "outputTokens": 664,
        "reasoningTokens": 62,
        "jevInputTokens": 10022,
        "mainSeconds": 25.028926460000456,
        "browserSeconds": 11.206691416999616,
        "mainCost": 0.0022015999999999997,
        "jevCost": 0.00042092399999999993,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 1.3552527156068805e-20
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "text": 2,
          "finish": 1
        },
        "totalCost": 0.00205228,
        "meanCost": 0.0006840933333333334,
        "costPerSuccess": 0.0006840933333333334,
        "meanNoCacheCost": 0.0024147333333333332,
        "meanSeconds": 31.051080333333346,
        "medianSeconds": 34.54045916699991,
        "p95Seconds": 40.866577653900066,
        "mainCalls": 18,
        "browserCalls": 15,
        "jevCalls": 0,
        "inputTokens": 32693,
        "cachedInputTokens": 29669,
        "cacheWriteTokens": 2970,
        "outputTokens": 588,
        "reasoningTokens": 59,
        "jevInputTokens": 0,
        "mainSeconds": 73.31032099999948,
        "browserSeconds": 19.842152624999578,
        "mainCost": 0.00205228,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.710505431213761e-20
      }
    },
    "openai/gpt-5.6-luna/preferences": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 2,
          "text": 1
        },
        "totalCost": 0.003034242,
        "meanCost": 0.001011414,
        "costPerSuccess": 0.001011414,
        "meanNoCacheCost": 0.002688904,
        "meanSeconds": 14.376142847333336,
        "medianSeconds": 14.284553208000027,
        "p95Seconds": 14.951632570799953,
        "mainCalls": 15,
        "browserCalls": 12,
        "jevCalls": 20,
        "inputTokens": 32814,
        "cachedInputTokens": 29004,
        "cacheWriteTokens": 3765,
        "outputTokens": 776,
        "reasoningTokens": 167,
        "jevInputTokens": 13636,
        "mainSeconds": 30.847894122999627,
        "browserSeconds": 12.280004625000526,
        "mainCost": 0.00246153,
        "jevCost": 0.0005727120000000001,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.710505431213761e-20
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "text": 1,
          "finish": 2
        },
        "totalCost": 0.00346441,
        "meanCost": 0.0011548033333333332,
        "costPerSuccess": 0.0011548033333333332,
        "meanNoCacheCost": 0.003623,
        "meanSeconds": 44.512347430666786,
        "medianSeconds": 39.613119833000006,
        "p95Seconds": 64.97025889610029,
        "mainCalls": 22,
        "browserCalls": 21,
        "jevCalls": 0,
        "inputTokens": 47373,
        "cachedInputTokens": 42478,
        "cacheWriteTokens": 4829,
        "outputTokens": 1162,
        "reasoningTokens": 216,
        "jevInputTokens": 0,
        "mainSeconds": 113.22988871099952,
        "browserSeconds": 20.306038664001274,
        "mainCost": 0.00346441,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.710505431213761e-20
      }
    },
    "openai/gpt-5.6-luna/catalog": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "text": 2,
          "finish": 1
        },
        "totalCost": 0.004592536,
        "meanCost": 0.0015308453333333335,
        "costPerSuccess": 0.0015308453333333335,
        "meanNoCacheCost": 0.0032508253333333333,
        "meanSeconds": 10.972320291333327,
        "medianSeconds": 10.45914683299989,
        "p95Seconds": 12.022725007700046,
        "mainCalls": 15,
        "browserCalls": 12,
        "jevCalls": 8,
        "inputTokens": 33572,
        "cachedInputTokens": 29723,
        "cacheWriteTokens": 3804,
        "outputTokens": 583,
        "reasoningTokens": 85,
        "jevInputTokens": 55678,
        "mainSeconds": 26.063001542000563,
        "browserSeconds": 6.853302124000038,
        "mainCost": 0.0022540600000000004,
        "jevCost": 0.0023384760000000004,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.710505431213761e-20
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "text": 3
        },
        "totalCost": 0.00475345,
        "meanCost": 0.0015844833333333334,
        "costPerSuccess": 0.0015844833333333334,
        "meanNoCacheCost": 0.004066466666666667,
        "meanSeconds": 15.088001916333448,
        "medianSeconds": 9.565196791000082,
        "p95Seconds": 25.503308004100276,
        "mainCalls": 15,
        "browserCalls": 12,
        "jevCalls": 0,
        "inputTokens": 58471,
        "cachedInputTokens": 45075,
        "cacheWriteTokens": 13351,
        "outputTokens": 421,
        "reasoningTokens": 52,
        "jevInputTokens": 0,
        "mainSeconds": 42.06229220699973,
        "browserSeconds": 3.2011422919996546,
        "mainCost": 0.00475345,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.168404344971009e-19
      }
    },
    "openai/gpt-5.6-luna/record": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.006363756000000001,
        "meanCost": 0.002121252,
        "costPerSuccess": 0.002121252,
        "meanNoCacheCost": 0.006238908666666667,
        "meanSeconds": 27.490157597333212,
        "medianSeconds": 27.548872166999615,
        "p95Seconds": 32.61936225449999,
        "mainCalls": 24,
        "browserCalls": 21,
        "jevCalls": 18,
        "inputTokens": 80758,
        "cachedInputTokens": 71249,
        "cacheWriteTokens": 9437,
        "outputTokens": 1231,
        "reasoningTokens": 298,
        "jevInputTokens": 25903,
        "mainSeconds": 70.2827664199998,
        "browserSeconds": 12.186821541000608,
        "mainCost": 0.0052758300000000004,
        "jevCost": 0.0010879260000000001,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 5.421010862427522e-20
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 2,
          "text": 1
        },
        "totalCost": 0.0053768900000000005,
        "meanCost": 0.0017922966666666668,
        "costPerSuccess": 0.0017922966666666668,
        "meanNoCacheCost": 0.006224266666666666,
        "meanSeconds": 39.72522416666666,
        "medianSeconds": 38.23989199999999,
        "p95Seconds": 55.708642212200004,
        "mainCalls": 26,
        "browserCalls": 23,
        "jevCalls": 0,
        "inputTokens": 87196,
        "cachedInputTokens": 76747,
        "cacheWriteTokens": 10371,
        "outputTokens": 1028,
        "reasoningTokens": 259,
        "jevInputTokens": 0,
        "mainSeconds": 96.69510874999978,
        "browserSeconds": 22.47946282799927,
        "mainCost": 0.0053768900000000005,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 5.421010862427522e-20
      }
    },
    "openai/gpt-5.6-luna/wizard": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 1,
          "text": 2
        },
        "totalCost": 0.003816778,
        "meanCost": 0.0012722593333333334,
        "costPerSuccess": 0.0012722593333333334,
        "meanNoCacheCost": 0.003836156,
        "meanSeconds": 16.734280527666694,
        "medianSeconds": 16.486127334000077,
        "p95Seconds": 17.33563875779999,
        "mainCalls": 21,
        "browserCalls": 18,
        "jevCalls": 21,
        "inputTokens": 48804,
        "cachedInputTokens": 44038,
        "cacheWriteTokens": 4703,
        "outputTokens": 926,
        "reasoningTokens": 103,
        "jevInputTokens": 15154,
        "mainSeconds": 36.41691466599924,
        "browserSeconds": 13.785112999000123,
        "mainCost": 0.0031803099999999996,
        "jevCost": 0.0006364680000000001,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.710505431213761e-20
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "text": 2,
          "finish": 1
        },
        "totalCost": 0.00368483,
        "meanCost": 0.0012282766666666667,
        "costPerSuccess": 0.0012282766666666667,
        "meanNoCacheCost": 0.004613533333333333,
        "meanSeconds": 47.35689013900005,
        "medianSeconds": 41.55109508300002,
        "p95Seconds": 73.9809521339001,
        "mainCalls": 29,
        "browserCalls": 28,
        "jevCalls": 0,
        "inputTokens": 63353,
        "cachedInputTokens": 57909,
        "cacheWriteTokens": 5357,
        "outputTokens": 975,
        "reasoningTokens": 100,
        "jevInputTokens": 0,
        "mainSeconds": 120.27740175099999,
        "browserSeconds": 21.792099208000874,
        "mainCost": 0.00368483,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.710505431213761e-20
      }
    },
    "openai/gpt-5.6-luna/validation": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 2,
          "text": 1
        },
        "totalCost": 0.0037631900000000005,
        "meanCost": 0.0012543966666666668,
        "costPerSuccess": 0.0012543966666666668,
        "meanNoCacheCost": 0.0038437966666666663,
        "meanSeconds": 20.065853666666644,
        "medianSeconds": 19.513116832999977,
        "p95Seconds": 22.634160871099965,
        "mainCalls": 21,
        "browserCalls": 18,
        "jevCalls": 20,
        "inputTokens": 48778,
        "cachedInputTokens": 44365,
        "cacheWriteTokens": 4350,
        "outputTokens": 997,
        "reasoningTokens": 191,
        "jevInputTokens": 13795,
        "mainSeconds": 44.14992437199952,
        "browserSeconds": 16.04685745799972,
        "mainCost": 0.0031837999999999997,
        "jevCost": 0.00057939,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.710505431213761e-20
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.00337845,
        "meanCost": 0.00112615,
        "costPerSuccess": 0.00112615,
        "meanNoCacheCost": 0.004169599999999999,
        "meanSeconds": 34.46446744433333,
        "medianSeconds": 29.771457999999985,
        "p95Seconds": 42.60850029969996,
        "mainCalls": 27,
        "browserCalls": 24,
        "jevCalls": 0,
        "inputTokens": 56628,
        "cachedInputTokens": 51990,
        "cacheWriteTokens": 4557,
        "outputTokens": 986,
        "reasoningTokens": 135,
        "jevInputTokens": 0,
        "mainSeconds": 65.28029820899997,
        "browserSeconds": 38.11211312799983,
        "mainCost": 0.00337845,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 1.3552527156068805e-20
      }
    },
    "openai/gpt-5.6-terra/contact": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.021225964,
        "meanCost": 0.007075321333333333,
        "costPerSuccess": 0.007075321333333333,
        "meanNoCacheCost": 0.02327165466666667,
        "meanSeconds": 12.832696805666666,
        "medianSeconds": 12.958504207999912,
        "p95Seconds": 13.391638470800036,
        "mainCalls": 15,
        "browserCalls": 12,
        "jevCalls": 14,
        "inputTokens": 30862,
        "cachedInputTokens": 27825,
        "cacheWriteTokens": 2992,
        "outputTokens": 637,
        "reasoningTokens": 24,
        "jevInputTokens": 10642,
        "mainSeconds": 27.502422333999537,
        "browserSeconds": 10.99512487499998,
        "mainCost": 0.020779,
        "jevCost": 0.000446964,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.168404344971009e-19
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.020606100000000002,
        "meanCost": 0.006868700000000001,
        "costPerSuccess": 0.006868700000000001,
        "meanNoCacheCost": 0.024164,
        "meanSeconds": 16.937283596666607,
        "medianSeconds": 16.749539915999865,
        "p95Seconds": 17.560529391299934,
        "mainCalls": 18,
        "browserCalls": 15,
        "jevCalls": 0,
        "inputTokens": 32640,
        "cachedInputTokens": 29643,
        "cacheWriteTokens": 2943,
        "outputTokens": 601,
        "reasoningTokens": 37,
        "jevInputTokens": 0,
        "mainSeconds": 30.96765783399995,
        "browserSeconds": 19.843376499000936,
        "mainCost": 0.020606100000000002,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.168404344971009e-19
      }
    },
    "openai/gpt-5.6-terra/preferences": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.024196074,
        "meanCost": 0.008065358,
        "costPerSuccess": 0.008065358,
        "meanNoCacheCost": 0.024846391333333332,
        "meanSeconds": 13.612103847000094,
        "medianSeconds": 13.359277875000146,
        "p95Seconds": 14.522121075000014,
        "mainCalls": 15,
        "browserCalls": 12,
        "jevCalls": 18,
        "inputTokens": 32492,
        "cachedInputTokens": 28942,
        "cacheWriteTokens": 3505,
        "outputTokens": 752,
        "reasoningTokens": 134,
        "jevInputTokens": 12647,
        "mainSeconds": 29.362962543000393,
        "browserSeconds": 11.472226253000272,
        "mainCost": 0.023664900000000003,
        "jevCost": 0.0005311739999999999,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 4.336808689942018e-19
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "text": 1,
          "finish": 2
        },
        "totalCost": 0.0256955,
        "meanCost": 0.008565166666666667,
        "costPerSuccess": 0.008565166666666667,
        "meanNoCacheCost": 0.028203333333333334,
        "meanSeconds": 19.02725159700002,
        "medianSeconds": 18.428410666000005,
        "p95Seconds": 20.456706654100064,
        "mainCalls": 19,
        "browserCalls": 16,
        "jevCalls": 0,
        "inputTokens": 37685,
        "cachedInputTokens": 33795,
        "cacheWriteTokens": 3833,
        "outputTokens": 770,
        "reasoningTokens": 106,
        "jevInputTokens": 0,
        "mainSeconds": 36.758816915999454,
        "browserSeconds": 20.32179100000035,
        "mainCost": 0.0256955,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 4.336808689942018e-19
      }
    },
    "openai/gpt-5.6-terra/catalog": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.02646457,
        "meanCost": 0.008821523333333333,
        "costPerSuccess": 0.008821523333333333,
        "meanNoCacheCost": 0.028846256666666664,
        "meanSeconds": 14.534026513666703,
        "medianSeconds": 14.477912917000008,
        "p95Seconds": 16.564978266100084,
        "mainCalls": 17,
        "browserCalls": 14,
        "jevCalls": 11,
        "inputTokens": 38161,
        "cachedInputTokens": 34404,
        "cacheWriteTokens": 3706,
        "outputTokens": 653,
        "reasoningTokens": 103,
        "jevInputTokens": 56685,
        "mainSeconds": 35.45364420699971,
        "browserSeconds": 8.14781395700021,
        "mainCost": 0.024083800000000002,
        "jevCost": 0.0023807700000000004,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 4.336808689942018e-19
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.047558699999999995,
        "meanCost": 0.0158529,
        "costPerSuccess": 0.0158529,
        "meanNoCacheCost": 0.040662,
        "meanSeconds": 9.44128840233333,
        "medianSeconds": 9.365804790999974,
        "p95Seconds": 10.126576441600012,
        "mainCalls": 15,
        "browserCalls": 12,
        "jevCalls": 0,
        "inputTokens": 58425,
        "cachedInputTokens": 45051,
        "cacheWriteTokens": 13329,
        "outputTokens": 428,
        "reasoningTokens": 52,
        "jevInputTokens": 0,
        "mainSeconds": 25.120342040000082,
        "browserSeconds": 3.203026459999979,
        "mainCost": 0.047558699999999995,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 1.734723475976807e-18
      }
    },
    "openai/gpt-5.6-terra/record": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.050892424,
        "meanCost": 0.01696414133333333,
        "costPerSuccess": 0.01696414133333333,
        "meanNoCacheCost": 0.057286708,
        "meanSeconds": 21.529106541333316,
        "medianSeconds": 21.77217433300009,
        "p95Seconds": 23.136044082699854,
        "mainCalls": 24,
        "browserCalls": 21,
        "jevCalls": 19,
        "inputTokens": 78867,
        "cachedInputTokens": 69724,
        "cacheWriteTokens": 9071,
        "outputTokens": 1084,
        "reasoningTokens": 221,
        "jevInputTokens": 26622,
        "mainSeconds": 51.747396916999534,
        "browserSeconds": 12.839067750000105,
        "mainCost": 0.049774299999999994,
        "jevCost": 0.001118124,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 8.673617379884035e-19
      },
      "playwright": {
        "trials": 3,
        "successes": 2,
        "naturalCompletions": 2,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.0725229,
        "meanCost": 0.0241743,
        "costPerSuccess": 0.03626145,
        "meanNoCacheCost": 0.09392333333333332,
        "meanSeconds": 34.04189495833332,
        "medianSeconds": 29.15017108300002,
        "p95Seconds": 47.34291313359998,
        "mainCalls": 33,
        "browserCalls": 30,
        "jevCalls": 0,
        "inputTokens": 133085,
        "cachedInputTokens": 119887,
        "cacheWriteTokens": 13099,
        "outputTokens": 1300,
        "reasoningTokens": 389,
        "jevInputTokens": 0,
        "mainSeconds": 73.17279320399953,
        "browserSeconds": 28.951602539000568,
        "mainCost": 0.0725229,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 8.673617379884035e-19
      }
    },
    "openai/gpt-5.6-terra/wizard": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.030717325999999996,
        "meanCost": 0.010239108666666665,
        "costPerSuccess": 0.010239108666666665,
        "meanNoCacheCost": 0.03592007533333333,
        "meanSeconds": 19.054701291999965,
        "medianSeconds": 16.826219749999932,
        "p95Seconds": 23.522506700599962,
        "mainCalls": 21,
        "browserCalls": 18,
        "jevCalls": 21,
        "inputTokens": 48644,
        "cachedInputTokens": 44058,
        "cacheWriteTokens": 4523,
        "outputTokens": 820,
        "reasoningTokens": 48,
        "jevInputTokens": 15053,
        "mainSeconds": 42.60833249799988,
        "browserSeconds": 14.55492691600017,
        "mainCost": 0.030085099999999997,
        "jevCost": 0.0006322260000000001,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.168404344971009e-19
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.0401377,
        "meanCost": 0.013379233333333332,
        "costPerSuccess": 0.013379233333333332,
        "meanNoCacheCost": 0.051028,
        "meanSeconds": 28.807272402999946,
        "medianSeconds": 29.177005542,
        "p95Seconds": 32.99080695449989,
        "mainCalls": 31,
        "browserCalls": 30,
        "jevCalls": 0,
        "inputTokens": 70422,
        "cachedInputTokens": 64396,
        "cacheWriteTokens": 5933,
        "outputTokens": 1020,
        "reasoningTokens": 172,
        "jevInputTokens": 0,
        "mainSeconds": 64.93075037599972,
        "browserSeconds": 21.489889168000257,
        "mainCost": 0.0401377,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 4.336808689942018e-19
      }
    },
    "openai/gpt-5.6-terra/validation": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.029818836,
        "meanCost": 0.009939612,
        "costPerSuccess": 0.009939612,
        "meanNoCacheCost": 0.035574178666666664,
        "meanSeconds": 20.639376402666677,
        "medianSeconds": 20.33190283299994,
        "p95Seconds": 23.607881233000068,
        "mainCalls": 21,
        "browserCalls": 18,
        "jevCalls": 18,
        "inputTokens": 47869,
        "cachedInputTokens": 43829,
        "cacheWriteTokens": 3977,
        "outputTokens": 873,
        "reasoningTokens": 107,
        "jevInputTokens": 12108,
        "mainSeconds": 46.186311624999625,
        "browserSeconds": 15.730792455999762,
        "mainCost": 0.029310300000000004,
        "jevCost": 0.000508536,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 4.336808689942018e-19
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 2,
          "text": 1
        },
        "totalCost": 0.034205,
        "meanCost": 0.011401666666666666,
        "costPerSuccess": 0.011401666666666666,
        "meanNoCacheCost": 0.04196666666666667,
        "meanSeconds": 28.849180458333265,
        "medianSeconds": 29.18637854199996,
        "p95Seconds": 30.586122216099874,
        "mainCalls": 27,
        "browserCalls": 24,
        "jevCalls": 0,
        "inputTokens": 56926,
        "cachedInputTokens": 52225,
        "cacheWriteTokens": 4620,
        "outputTokens": 1004,
        "reasoningTokens": 107,
        "jevInputTokens": 0,
        "mainSeconds": 48.51086929200007,
        "browserSeconds": 38.0355659180003,
        "mainCost": 0.034205,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 2.168404344971009e-19
      }
    },
    "openai/gpt-6-astra/contact": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.085200342,
        "meanCost": 0.028400114,
        "costPerSuccess": 0.028400114,
        "meanNoCacheCost": 0.10797211400000001,
        "meanSeconds": 17.031276111000043,
        "medianSeconds": 16.343753125000045,
        "p95Seconds": 18.815298925000004,
        "mainCalls": 15,
        "browserCalls": 12,
        "jevCalls": 9,
        "inputTokens": 29950,
        "cachedInputTokens": 27259,
        "cacheWriteTokens": 2646,
        "outputTokens": 484,
        "reasoningTokens": 0,
        "jevInputTokens": 5151,
        "mainSeconds": 41.39242850099958,
        "browserSeconds": 9.70082825099968,
        "mainCost": 0.084984,
        "jevCost": 0.00021634199999999998,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 8.673617379884035e-19
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.09304950000000001,
        "meanCost": 0.031016500000000002,
        "costPerSuccess": 0.031016500000000002,
        "meanNoCacheCost": 0.11703000000000001,
        "meanSeconds": 26.695035833333307,
        "medianSeconds": 23.208173707999986,
        "p95Seconds": 33.215256533299964,
        "mainCalls": 18,
        "browserCalls": 15,
        "jevCalls": 0,
        "inputTokens": 32409,
        "cachedInputTokens": 29472,
        "cacheWriteTokens": 2883,
        "outputTokens": 540,
        "reasoningTokens": 0,
        "jevInputTokens": 0,
        "mainSeconds": 60.172248208999754,
        "browserSeconds": 19.912301582999994,
        "mainCost": 0.09304950000000001,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 8.673617379884035e-19
      }
    },
    "openai/gpt-6-astra/preferences": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.09933949600000001,
        "meanCost": 0.03311316533333334,
        "costPerSuccess": 0.03311316533333334,
        "meanNoCacheCost": 0.11528266533333335,
        "meanSeconds": 16.74574218033332,
        "medianSeconds": 16.76196037499979,
        "p95Seconds": 18.167196561900063,
        "mainCalls": 15,
        "browserCalls": 12,
        "jevCalls": 15,
        "inputTokens": 31801,
        "cachedInputTokens": 28339,
        "cacheWriteTokens": 3417,
        "outputTokens": 549,
        "reasoningTokens": 0,
        "jevInputTokens": 9238,
        "mainSeconds": 40.26919420699985,
        "browserSeconds": 9.967296584000344,
        "mainCost": 0.09895150000000001,
        "jevCost": 0.00038799600000000004,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 0
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.10640250000000001,
        "meanCost": 0.035467500000000006,
        "costPerSuccess": 0.035467500000000006,
        "meanNoCacheCost": 0.12492000000000002,
        "meanSeconds": 40.45738736100006,
        "medianSeconds": 38.40649195800023,
        "p95Seconds": 49.35022570830004,
        "mainCalls": 18,
        "browserCalls": 15,
        "jevCalls": 0,
        "inputTokens": 34161,
        "cachedInputTokens": 30750,
        "cacheWriteTokens": 3357,
        "outputTokens": 663,
        "reasoningTokens": 0,
        "jevInputTokens": 0,
        "mainSeconds": 101.323747706,
        "browserSeconds": 20.047648666998835,
        "mainCost": 0.10640250000000001,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 8.673617379884035e-19
      }
    },
    "openai/gpt-6-astra/catalog": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.157326954,
        "meanCost": 0.052442317999999995,
        "costPerSuccess": 0.052442317999999995,
        "meanNoCacheCost": 0.15795948466666668,
        "meanSeconds": 15.35878511099986,
        "medianSeconds": 15.28494487499958,
        "p95Seconds": 16.722793499699957,
        "mainCalls": 15,
        "browserCalls": 12,
        "jevCalls": 6,
        "inputTokens": 45055,
        "cachedInputTokens": 37311,
        "cacheWriteTokens": 7699,
        "outputTokens": 446,
        "reasoningTokens": 0,
        "jevInputTokens": 24487,
        "mainSeconds": 39.704114375000124,
        "browserSeconds": 6.371634331000269,
        "mainCost": 0.1562985,
        "jevCost": 0.001028454,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 3.469446951953614e-18
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.22666550000000002,
        "meanCost": 0.07555516666666667,
        "costPerSuccess": 0.07555516666666667,
        "meanNoCacheCost": 0.19922333333333334,
        "meanSeconds": 21.004387500333376,
        "medianSeconds": 15.683617166999989,
        "p95Seconds": 31.23401662949997,
        "mainCalls": 15,
        "browserCalls": 12,
        "jevCalls": 0,
        "inputTokens": 57967,
        "cachedInputTokens": 44853,
        "cacheWriteTokens": 13069,
        "outputTokens": 360,
        "reasoningTokens": 0,
        "jevInputTokens": 0,
        "mainSeconds": 59.66887670899943,
        "browserSeconds": 3.343734583000187,
        "mainCost": 0.22666550000000002,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 1.734723475976807e-18
      }
    },
    "openai/gpt-6-astra/record": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.20621062799999998,
        "meanCost": 0.06873687599999999,
        "costPerSuccess": 0.06873687599999999,
        "meanNoCacheCost": 0.2360460426666667,
        "meanSeconds": 26.272817875333363,
        "medianSeconds": 23.937307709000073,
        "p95Seconds": 34.324863133099996,
        "mainCalls": 21,
        "browserCalls": 18,
        "jevCalls": 10,
        "inputTokens": 67725,
        "cachedInputTokens": 58355,
        "cacheWriteTokens": 9307,
        "outputTokens": 604,
        "reasoningTokens": 0,
        "jevInputTokens": 16384,
        "mainSeconds": 68.97801554200052,
        "browserSeconds": 9.839107126999647,
        "mainCost": 0.20552249999999997,
        "jevCost": 0.0006881280000000001,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 3.469446951953614e-18
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.17227549999999997,
        "meanCost": 0.05742516666666666,
        "costPerSuccess": 0.05742516666666666,
        "meanNoCacheCost": 0.20832000000000003,
        "meanSeconds": 36.548322416999916,
        "medianSeconds": 30.228663333999926,
        "p95Seconds": 48.03017395809995,
        "mainCalls": 24,
        "browserCalls": 21,
        "jevCalls": 0,
        "inputTokens": 59201,
        "cachedInputTokens": 52218,
        "cacheWriteTokens": 6911,
        "outputTokens": 659,
        "reasoningTokens": 0,
        "jevInputTokens": 0,
        "mainSeconds": 87.99744454200008,
        "browserSeconds": 21.646579168000258,
        "mainCost": 0.17227549999999997,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 1.734723475976807e-18
      }
    },
    "openai/gpt-6-astra/wizard": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.127075612,
        "meanCost": 0.042358537333333335,
        "costPerSuccess": 0.042358537333333335,
        "meanNoCacheCost": 0.16651353733333338,
        "meanSeconds": 23.13398497199997,
        "medianSeconds": 22.28086220799992,
        "p95Seconds": 26.07911424580002,
        "mainCalls": 21,
        "browserCalls": 18,
        "jevCalls": 15,
        "inputTokens": 46738,
        "cachedInputTokens": 42535,
        "cacheWriteTokens": 4140,
        "outputTokens": 636,
        "reasoningTokens": 0,
        "jevInputTokens": 8586,
        "mainSeconds": 57.17539270999981,
        "browserSeconds": 12.225541372999782,
        "mainCost": 0.12671500000000002,
        "jevCost": 0.00036061200000000006,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 1.734723475976807e-18
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.15425900000000003,
        "meanCost": 0.051419666666666676,
        "costPerSuccess": 0.051419666666666676,
        "meanNoCacheCost": 0.22293666666666667,
        "meanSeconds": 74.47394709733338,
        "medianSeconds": 77.35744954199996,
        "p95Seconds": 87.92721499170018,
        "mainCalls": 30,
        "browserCalls": 27,
        "jevCalls": 0,
        "inputTokens": 63411,
        "cachedInputTokens": 58509,
        "cacheWriteTokens": 4812,
        "outputTokens": 694,
        "reasoningTokens": 0,
        "jevInputTokens": 0,
        "mainSeconds": 201.84886316299998,
        "browserSeconds": 21.571904413998592,
        "mainCost": 0.15425900000000003,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 8.673617379884035e-19
      }
    },
    "openai/gpt-6-astra/validation": {
      "plain": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.12723922000000001,
        "meanCost": 0.042413073333333336,
        "costPerSuccess": 0.042413073333333336,
        "meanNoCacheCost": 0.16775140666666669,
        "meanSeconds": 23.866823986333312,
        "medianSeconds": 24.030316082999985,
        "p95Seconds": 24.08836199609993,
        "mainCalls": 21,
        "browserCalls": 18,
        "jevCalls": 15,
        "inputTokens": 47043,
        "cachedInputTokens": 42910,
        "cacheWriteTokens": 4070,
        "outputTokens": 649,
        "reasoningTokens": 0,
        "jevInputTokens": 8910,
        "mainSeconds": 56.68429779399972,
        "browserSeconds": 14.915486581999517,
        "mainCost": 0.126865,
        "jevCost": 0.00037422000000000003,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 1.734723475976807e-18
      },
      "playwright": {
        "trials": 3,
        "successes": 3,
        "naturalCompletions": 3,
        "terminations": {
          "finish": 3
        },
        "totalCost": 0.14186500000000002,
        "meanCost": 0.04728833333333334,
        "costPerSuccess": 0.04728833333333334,
        "meanNoCacheCost": 0.19721333333333335,
        "meanSeconds": 51.40822626333331,
        "medianSeconds": 42.32796716599993,
        "p95Seconds": 74.40402514129997,
        "mainCalls": 27,
        "browserCalls": 24,
        "jevCalls": 0,
        "inputTokens": 55599,
        "cachedInputTokens": 51180,
        "cacheWriteTokens": 4338,
        "outputTokens": 713,
        "reasoningTokens": 0,
        "jevInputTokens": 0,
        "mainSeconds": 115.7796419129999,
        "browserSeconds": 38.44402937599982,
        "mainCost": 0.14186500000000002,
        "jevCost": 0,
        "incompleteUsageTrials": 0,
        "outputLimitGenerations": 0,
        "nonObjectToolArguments": 0,
        "maxReportedCostDrift": 8.673617379884035e-19
      }
    }
  },
  "pairedSuccessful": {
    "openai/gpt-5.6-luna": {
      "costRatio": 1.065288232525228,
      "costRatio95": [
        0.9667475996549272,
        1.1626448769783335
      ],
      "timeRatio": 0.4793518399286593,
      "timeRatio95": [
        0.36807094754300174,
        0.6223587345260233
      ]
    },
    "openai/gpt-5.6-terra": {
      "costRatio": 0.8006315526309262,
      "costRatio95": [
        0.677053734658419,
        0.9308514793839954
      ],
      "timeRatio": 0.7827596891601768,
      "timeRatio95": [
        0.6951435369248643,
        0.9768915909397634
      ]
    },
    "openai/gpt-6-astra": {
      "costRatio": 0.8970117415320222,
      "costRatio95": [
        0.7667083524792284,
        1.057576591363709
      ],
      "timeRatio": 0.488490147243411,
      "timeRatio95": [
        0.380147942196216,
        0.6470593800157557
      ]
    }
  },
  "noOutputCapSensitivity": {
    "openai/gpt-5.6-luna": {
      "trials": 36,
      "costRatio": 1.065288232525228,
      "costRatio95": [
        0.9667475996549272,
        1.1626448769783335
      ],
      "timeRatio": 0.4793518399286593,
      "timeRatio95": [
        0.36807094754300174,
        0.6223587345260233
      ]
    },
    "openai/gpt-5.6-terra": {
      "trials": 36,
      "costRatio": 0.7615100577046343,
      "costRatio95": [
        0.6635927849383548,
        0.9067622385455407
      ],
      "timeRatio": 0.7454332741815841,
      "timeRatio95": [
        0.6619233274957169,
        0.94343589829332
      ]
    },
    "openai/gpt-6-astra": {
      "trials": 36,
      "costRatio": 0.8970117415320222,
      "costRatio95": [
        0.7667083524792284,
        1.057576591363709
      ],
      "timeRatio": 0.488490147243411,
      "timeRatio95": [
        0.380147942196216,
        0.6470593800157557
      ]
    }
  },
  "pairedSuccessfulCompleteNatural": {
    "openai/gpt-5.6-luna": {
      "pairs": 18,
      "costRatio": 1.065288232525228,
      "costRatio95": [
        0.9667475996549272,
        1.1626448769783335
      ],
      "timeRatio": 0.4793518399286593,
      "timeRatio95": [
        0.36807094754300174,
        0.6223587345260233
      ]
    },
    "openai/gpt-5.6-terra": {
      "pairs": 17,
      "costRatio": 0.8006315526309262,
      "costRatio95": [
        0.677053734658419,
        0.9308514793839954
      ],
      "timeRatio": 0.7827596891601768,
      "timeRatio95": [
        0.6951435369248643,
        0.9768915909397634
      ]
    },
    "openai/gpt-6-astra": {
      "pairs": 18,
      "costRatio": 0.8970117415320222,
      "costRatio95": [
        0.7667083524792284,
        1.057576591363709
      ],
      "timeRatio": 0.488490147243411,
      "timeRatio95": [
        0.380147942196216,
        0.6470593800157557
      ]
    }
  }
}