{
  "status": "published",
  "generatedAt": "2026-08-09T07:03:49Z",
  "commit": "3e065a78",
  "suiteVersion": "1.0",
  "sourceUrl": "https://fuaran-ui.io/eval/README.md",
  "cohort": "since-20260730T003649Z",
  "summary": {
    "totalPrompts": 1235,
    "tasks": 57,
    "passed": 1015,
    "passRate": 0.8218623481781376,
    "passRateCi95": 0.05390585129480917,
    "parsePassRate": 0.8582995951417004,
    "providersEvaluated": 7,
    "reasoningShare": 0.5862801281812626,
    "costPerCorrectUsd": 0.08620864330049272,
    "costPerCorrectCachedUsd": 0.08432159527093605
  },
  "providers": [
    {
      "name": "Google Gemini",
      "model": "gemini-3.6-flash",
      "tier": "headline",
      "budgetPick": true,
      "passed": 156,
      "total": 174,
      "passRate": 0.896551724137931,
      "passRateCi95": 0.06125654064999244,
      "parsePassRate": 0.9310344827586207,
      "meanOutputTokens": 1730.4770114942528,
      "costUsd": 10.8836751,
      "reasoningShare": 0.7330003334137309,
      "reasoningSplitSource": "native",
      "costPerCorrectUsd": 0.06976714807692308,
      "costPerCorrectCachedUsd": 0.06976714807692308
    },
    {
      "name": "OpenAI GPT",
      "model": "gpt-5.6-sol@low",
      "tier": "headline",
      "budgetPick": false,
      "passed": 148,
      "total": 171,
      "passRate": 0.8654970760233918,
      "passRateCi95": 0.07287929431105274,
      "parsePassRate": 0.9064327485380117,
      "meanOutputTokens": 1259.0116959064328,
      "costUsd": 9.380124,
      "reasoningShare": 0.19026700115466058,
      "reasoningSplitSource": "sdk",
      "costPerCorrectUsd": 0.0633792162162162,
      "costPerCorrectCachedUsd": 0.0633792162162162
    },
    {
      "name": "xAI Grok",
      "model": "grok-4.5",
      "tier": "headline",
      "budgetPick": true,
      "passed": 162,
      "total": 192,
      "passRate": 0.84375,
      "passRateCi95": 0.07129913627442923,
      "parsePassRate": 0.8645833333333334,
      "meanOutputTokens": 1582.5885416666667,
      "costUsd": 7.5064264,
      "reasoningShare": 0.4653417088963964,
      "reasoningSplitSource": "sdk",
      "costPerCorrectUsd": 0.046335965432098766,
      "costPerCorrectCachedUsd": 0.046335965432098766
    },
    {
      "name": "Claude (Opus)",
      "model": "claude-fable-5@low",
      "tier": "headline",
      "budgetPick": false,
      "passed": 143,
      "total": 171,
      "passRate": 0.8362573099415205,
      "passRateCi95": 0.08516981229021083,
      "parsePassRate": 0.8596491228070176,
      "meanOutputTokens": 1718.2280701754387,
      "costUsd": 23.011419,
      "reasoningShare": 0.16979506031211414,
      "reasoningSplitSource": "derived",
      "costPerCorrectUsd": 0.16091901398601405,
      "costPerCorrectCachedUsd": 0.15360823076923083
    },
    {
      "name": "Google Gemini",
      "model": "gemini-3.1-pro-preview",
      "tier": "budget",
      "budgetPick": false,
      "passed": 140,
      "total": 171,
      "passRate": 0.8187134502923976,
      "passRateCi95": 0.08508713651053411,
      "parsePassRate": 0.8421052631578947,
      "meanOutputTokens": 1587.7251461988303,
      "costUsd": 21.8689392,
      "reasoningShare": 0.827090572363125,
      "reasoningSplitSource": "native",
      "costPerCorrectUsd": 0.15620670857142854,
      "costPerCorrectCachedUsd": 0.15620670857142854
    },
    {
      "name": "Claude (Opus)",
      "model": "claude-opus-4-8@low",
      "tier": "budget",
      "budgetPick": true,
      "passed": 145,
      "total": 180,
      "passRate": 0.8055555555555556,
      "passRateCi95": 0.07958804366465794,
      "parsePassRate": 0.8833333333333333,
      "meanOutputTokens": 1626.5055555555555,
      "costUsd": 11.05585175,
      "reasoningShare": 0.08672310744544127,
      "reasoningSplitSource": "derived",
      "costPerCorrectUsd": 0.07624725344827585,
      "costPerCorrectCachedUsd": 0.07024786206896551
    },
    {
      "name": "OpenAI GPT",
      "model": "gpt-5.6-terra@low",
      "tier": "budget",
      "budgetPick": true,
      "passed": 121,
      "total": 176,
      "passRate": 0.6875,
      "passRateCi95": 0.09236045985947963,
      "parsePassRate": 0.7215909090909091,
      "meanOutputTokens": 818.7556818181819,
      "costUsd": 3.7953375,
      "reasoningShare": 0.2737284035239804,
      "reasoningSplitSource": "sdk",
      "costPerCorrectUsd": 0.03136642561983471,
      "costPerCorrectCachedUsd": 0.03136642561983471
    }
  ],
  "compactness": [
    {
      "complexity": "all",
      "baseline": "JSX \u002B shadcn/ui",
      "baselineId": "jsx_shadcn",
      "fuaranTokens": 1464.123836336336,
      "baselineTokens": 2047.6405405405412,
      "ratio": 0.7150297170566045,
      "fuaranBilledTokens": 3408.637975475476,
      "baselineBilledTokens": 2893.319744744746,
      "billedRatio": 1.178106215763648,
      "fuaranColdCallUsd": 0.06648314943881382,
      "baselineColdCallUsd": 0.046039381876876874,
      "coldCallBillRatio": 1.4440495664474757,
      "fuaranCachedTurnUsd": 0.06514184977352351,
      "baselineCachedTurnUsd": 0.04595064539039039,
      "cachedTurnBillRatio": 1.417648201022799,
      "pairs": 222,
      "tasks": 57
    },
    {
      "complexity": "all",
      "baseline": "Feliz (bare)",
      "baselineId": "feliz_bare",
      "fuaranTokens": 1366.3665146489775,
      "baselineTokens": 2021.809203980099,
      "ratio": 0.67581377706619,
      "fuaranBilledTokens": 3330.749571586513,
      "baselineBilledTokens": 3269.9256218905484,
      "billedRatio": 1.0186010193286288,
      "fuaranColdCallUsd": 0.06492505672125483,
      "baselineColdCallUsd": 0.05229804233001658,
      "coldCallBillRatio": 1.241443347182251,
      "fuaranCachedTurnUsd": 0.06344362127003868,
      "baselineCachedTurnUsd": 0.05225853611111111,
      "cachedTurnBillRatio": 1.2140336486874805,
      "pairs": 201,
      "tasks": 54
    },
    {
      "complexity": "all",
      "baseline": "Dash (Python)",
      "baselineId": "dash_py",
      "fuaranTokens": 1800.1175105485238,
      "baselineTokens": 2806.6561181434604,
      "ratio": 0.6413744451668917,
      "fuaranBilledTokens": 3733.8468354430383,
      "baselineBilledTokens": 4191.801687763715,
      "billedRatio": 0.8907498764415569,
      "fuaranColdCallUsd": 0.07361931185654008,
      "baselineColdCallUsd": 0.05997801751054853,
      "coldCallBillRatio": 1.2274382334093055,
      "fuaranCachedTurnUsd": 0.07116958137130802,
      "baselineCachedTurnUsd": 0.05981352383966245,
      "cachedTurnBillRatio": 1.1898576910814833,
      "pairs": 79,
      "tasks": 20
    },
    {
      "complexity": "trivial",
      "baseline": "JSX \u002B shadcn/ui",
      "baselineId": "jsx_shadcn",
      "fuaranTokens": 289.61979166666674,
      "baselineTokens": 879.071875,
      "ratio": 0.32946087789086276,
      "fuaranBilledTokens": 1026.0447916666667,
      "baselineBilledTokens": 1460.63125,
      "billedRatio": 0.7024666846383485,
      "fuaranColdCallUsd": 0.029839614583333337,
      "baselineColdCallUsd": 0.021047369791666666,
      "coldCallBillRatio": 1.4177360344164145,
      "fuaranCachedTurnUsd": 0.029197920572916668,
      "baselineCachedTurnUsd": 0.021047369791666666,
      "cachedTurnBillRatio": 1.3872479488851415,
      "pairs": 32,
      "tasks": 8
    },
    {
      "complexity": "trivial",
      "baseline": "Feliz (bare)",
      "baselineId": "feliz_bare",
      "fuaranTokens": 289.61979166666674,
      "baselineTokens": 647.4677083333332,
      "ratio": 0.4473115615482756,
      "fuaranBilledTokens": 1026.0447916666667,
      "baselineBilledTokens": 1261.1619791666667,
      "billedRatio": 0.8135709834391317,
      "fuaranColdCallUsd": 0.029839614583333337,
      "baselineColdCallUsd": 0.019517895833333333,
      "coldCallBillRatio": 1.5288335811472167,
      "fuaranCachedTurnUsd": 0.029197920572916668,
      "baselineCachedTurnUsd": 0.019517895833333333,
      "cachedTurnBillRatio": 1.4959563685677355,
      "pairs": 32,
      "tasks": 8
    },
    {
      "complexity": "trivial",
      "baseline": "Dash (Python)",
      "baselineId": "dash_py",
      "fuaranTokens": 371.96666666666664,
      "baselineTokens": 1541.1458333333335,
      "ratio": 0.24135721527543086,
      "fuaranBilledTokens": 1046.0041666666666,
      "baselineBilledTokens": 2226.5,
      "billedRatio": 0.4697975147840407,
      "fuaranColdCallUsd": 0.029372919166666667,
      "baselineColdCallUsd": 0.031276986458333336,
      "coldCallBillRatio": 0.9391224185167828,
      "fuaranCachedTurnUsd": 0.029372919166666667,
      "baselineCachedTurnUsd": 0.031276986458333336,
      "cachedTurnBillRatio": 0.9391224185167828,
      "pairs": 8,
      "tasks": 2
    },
    {
      "complexity": "medium",
      "baseline": "JSX \u002B shadcn/ui",
      "baselineId": "jsx_shadcn",
      "fuaranTokens": 1048.1814814814813,
      "baselineTokens": 1522.7377104377101,
      "ratio": 0.6883532694413814,
      "fuaranBilledTokens": 3044.158080808082,
      "baselineBilledTokens": 2250.690235690236,
      "billedRatio": 1.3525442251161264,
      "fuaranColdCallUsd": 0.05760473558922559,
      "baselineColdCallUsd": 0.035700897727272723,
      "coldCallBillRatio": 1.613537453015363,
      "fuaranCachedTurnUsd": 0.05716473474747476,
      "baselineCachedTurnUsd": 0.035700897727272723,
      "cachedTurnBillRatio": 1.6012128093856117,
      "pairs": 99,
      "tasks": 26
    },
    {
      "complexity": "medium",
      "baseline": "Feliz (bare)",
      "baselineId": "feliz_bare",
      "fuaranTokens": 1049.958422939068,
      "baselineTokens": 1613.1709677419353,
      "ratio": 0.6508661784366018,
      "fuaranBilledTokens": 3105.2569892473134,
      "baselineBilledTokens": 2817.6951612903217,
      "billedRatio": 1.1020556914415494,
      "fuaranColdCallUsd": 0.058123608046595006,
      "baselineColdCallUsd": 0.04446793969534052,
      "coldCallBillRatio": 1.307090196775754,
      "fuaranCachedTurnUsd": 0.05765522005376346,
      "baselineCachedTurnUsd": 0.04443947822580647,
      "cachedTurnBillRatio": 1.2973874211754914,
      "pairs": 93,
      "tasks": 25
    },
    {
      "complexity": "medium",
      "baseline": "Dash (Python)",
      "baselineId": "dash_py",
      "fuaranTokens": 984.9938271604941,
      "baselineTokens": 2089.777777777778,
      "ratio": 0.47133902830946656,
      "fuaranBilledTokens": 2807.1604938271603,
      "baselineBilledTokens": 3429.3518518518513,
      "billedRatio": 0.8185688185688187,
      "fuaranColdCallUsd": 0.055973231172839505,
      "baselineColdCallUsd": 0.04587168209876544,
      "coldCallBillRatio": 1.2202131818999926,
      "fuaranCachedTurnUsd": 0.05435989475308642,
      "baselineCachedTurnUsd": 0.04587168209876544,
      "cachedTurnBillRatio": 1.1850425418463002,
      "pairs": 27,
      "tasks": 7
    },
    {
      "complexity": "compound",
      "baseline": "JSX \u002B shadcn/ui",
      "baselineId": "jsx_shadcn",
      "fuaranTokens": 2329.644963369964,
      "baselineTokens": 3029.6139194139205,
      "ratio": 0.7689577039640201,
      "fuaranBilledTokens": 4642.9950244200245,
      "baselineBilledTokens": 4096.246703296704,
      "billedRatio": 1.1334754375714946,
      "fuaranColdCallUsd": 0.08902772181776558,
      "baselineColdCallUsd": 0.06607514360805858,
      "coldCallBillRatio": 1.3473708410814211,
      "fuaranCachedTurnUsd": 0.08645987309218563,
      "baselineCachedTurnUsd": 0.06585866558608056,
      "cachedTurnBillRatio": 1.3128093671922074,
      "pairs": 91,
      "tasks": 23
    },
    {
      "complexity": "compound",
      "baseline": "Feliz (bare)",
      "baselineId": "feliz_bare",
      "fuaranTokens": 2206.917141812866,
      "baselineTokens": 3100.5234649122813,
      "ratio": 0.7117885630564976,
      "fuaranBilledTokens": 4577.083296783625,
      "baselineBilledTokens": 4669.108114035087,
      "billedRatio": 0.9802907075604352,
      "fuaranColdCallUsd": 0.088020699815424,
      "baselineColdCallUsd": 0.07568178223684212,
      "coldCallBillRatio": 1.1630368262201847,
      "fuaranCachedTurnUsd": 0.08494603884137429,
      "baselineCachedTurnUsd": 0.07561212653508774,
      "cachedTurnBillRatio": 1.1234446474925575,
      "pairs": 76,
      "tasks": 21
    },
    {
      "complexity": "compound",
      "baseline": "Dash (Python)",
      "baselineId": "dash_py",
      "fuaranTokens": 2559.9708333333338,
      "baselineTokens": 3476.6515151515136,
      "ratio": 0.7363323077325366,
      "fuaranBilledTokens": 4791.193939393939,
      "baselineBilledTokens": 5016.996212121211,
      "billedRatio": 0.95499253673309,
      "fuaranColdCallUsd": 0.09249238731060606,
      "baselineColdCallUsd": 0.07385254715909091,
      "coldCallBillRatio": 1.2523926508771834,
      "fuaranCachedTurnUsd": 0.08908400946969695,
      "baselineCachedTurnUsd": 0.07355720625,
      "cachedTurnBillRatio": 1.2110847327034928,
      "pairs": 44,
      "tasks": 11
    }
  ],
  "costPerCorrect": [
    {
      "condition": "Fuaran",
      "conditionId": "fuaran",
      "runs": 1235,
      "passRate": 0.8218623481781376,
      "meanColdCallUsd": 0.07085163801619442,
      "meanCachedTurnUsd": 0.06930074429149805,
      "coldUsdPerCorrect": 0.08620864330049272,
      "cachedUsdPerCorrect": 0.08432159527093605
    },
    {
      "condition": "Fuaran",
      "conditionId": "fuaran",
      "provider": "Claude (Opus)",
      "providerModelId": "claude-fable-5@low",
      "runs": 171,
      "passRate": 0.8362573099415205,
      "meanColdCallUsd": 0.134569701754386,
      "meanCachedTurnUsd": 0.12845600584795328,
      "coldUsdPerCorrect": 0.16091901398601405,
      "cachedUsdPerCorrect": 0.15360823076923083
    },
    {
      "condition": "Fuaran",
      "conditionId": "fuaran",
      "provider": "Claude (Opus)",
      "providerModelId": "claude-opus-4-8@low",
      "runs": 180,
      "passRate": 0.8055555555555556,
      "meanColdCallUsd": 0.0614213986111111,
      "meanCachedTurnUsd": 0.056588555555555545,
      "coldUsdPerCorrect": 0.07624725344827585,
      "cachedUsdPerCorrect": 0.07024786206896551
    },
    {
      "condition": "Fuaran",
      "conditionId": "fuaran",
      "provider": "Google Gemini",
      "providerModelId": "gemini-3.1-pro-preview",
      "runs": 171,
      "passRate": 0.8187134502923976,
      "meanColdCallUsd": 0.1278885333333333,
      "meanCachedTurnUsd": 0.1278885333333333,
      "coldUsdPerCorrect": 0.15620670857142854,
      "cachedUsdPerCorrect": 0.15620670857142854
    },
    {
      "condition": "Fuaran",
      "conditionId": "fuaran",
      "provider": "Google Gemini",
      "providerModelId": "gemini-3.6-flash",
      "runs": 174,
      "passRate": 0.896551724137931,
      "meanColdCallUsd": 0.06254985689655174,
      "meanCachedTurnUsd": 0.06254985689655174,
      "coldUsdPerCorrect": 0.0697671480769231,
      "cachedUsdPerCorrect": 0.0697671480769231
    },
    {
      "condition": "Fuaran",
      "conditionId": "fuaran",
      "provider": "OpenAI GPT",
      "providerModelId": "gpt-5.6-sol@low",
      "runs": 171,
      "passRate": 0.8654970760233918,
      "meanColdCallUsd": 0.05485452631578946,
      "meanCachedTurnUsd": 0.05485452631578946,
      "coldUsdPerCorrect": 0.0633792162162162,
      "cachedUsdPerCorrect": 0.0633792162162162
    },
    {
      "condition": "Fuaran",
      "conditionId": "fuaran",
      "provider": "OpenAI GPT",
      "providerModelId": "gpt-5.6-terra@low",
      "runs": 176,
      "passRate": 0.6875,
      "meanColdCallUsd": 0.021564417613636363,
      "meanCachedTurnUsd": 0.021564417613636363,
      "coldUsdPerCorrect": 0.03136642561983471,
      "cachedUsdPerCorrect": 0.03136642561983471
    },
    {
      "condition": "Fuaran",
      "conditionId": "fuaran",
      "provider": "xAI Grok",
      "providerModelId": "grok-4.5",
      "runs": 192,
      "passRate": 0.84375,
      "meanColdCallUsd": 0.039095970833333334,
      "meanCachedTurnUsd": 0.039095970833333334,
      "coldUsdPerCorrect": 0.046335965432098766,
      "cachedUsdPerCorrect": 0.046335965432098766
    },
    {
      "condition": "JSX \u002B shadcn/ui",
      "conditionId": "jsx_shadcn",
      "runs": 1200,
      "passRate": 0.9841666666666666,
      "meanColdCallUsd": 0.049855587666666715,
      "meanCachedTurnUsd": 0.049757090166666705,
      "coldUsdPerCorrect": 0.050657667400508095,
      "cachedUsdPerCorrect": 0.05055758526672316
    },
    {
      "condition": "JSX \u002B shadcn/ui",
      "conditionId": "jsx_shadcn",
      "provider": "Claude (Opus)",
      "providerModelId": "claude-fable-5@low",
      "runs": 171,
      "passRate": 0.9883040935672515,
      "meanColdCallUsd": 0.09343629239766085,
      "meanCachedTurnUsd": 0.09312908771929827,
      "coldUsdPerCorrect": 0.09454204733727814,
      "cachedUsdPerCorrect": 0.09423120710059174
    },
    {
      "condition": "JSX \u002B shadcn/ui",
      "conditionId": "jsx_shadcn",
      "provider": "Claude (Opus)",
      "providerModelId": "claude-opus-4-8@low",
      "runs": 171,
      "passRate": 0.9824561403508771,
      "meanColdCallUsd": 0.0488384619883041,
      "meanCachedTurnUsd": 0.04845445614035088,
      "coldUsdPerCorrect": 0.04971057738095239,
      "cachedUsdPerCorrect": 0.049319714285714286
    },
    {
      "condition": "JSX \u002B shadcn/ui",
      "conditionId": "jsx_shadcn",
      "provider": "Google Gemini",
      "providerModelId": "gemini-3.1-pro-preview",
      "runs": 171,
      "passRate": 0.9707602339181286,
      "meanColdCallUsd": 0.05924340350877194,
      "meanCachedTurnUsd": 0.05924340350877194,
      "coldUsdPerCorrect": 0.06102784337349399,
      "cachedUsdPerCorrect": 0.06102784337349399
    },
    {
      "condition": "JSX \u002B shadcn/ui",
      "conditionId": "jsx_shadcn",
      "provider": "Google Gemini",
      "providerModelId": "gemini-3.6-flash",
      "runs": 165,
      "passRate": 0.9939393939393939,
      "meanColdCallUsd": 0.06243571818181819,
      "meanCachedTurnUsd": 0.06243571818181819,
      "coldUsdPerCorrect": 0.06281642378048781,
      "cachedUsdPerCorrect": 0.06281642378048781
    },
    {
      "condition": "JSX \u002B shadcn/ui",
      "conditionId": "jsx_shadcn",
      "provider": "OpenAI GPT",
      "providerModelId": "gpt-5.6-sol@low",
      "runs": 171,
      "passRate": 0.9766081871345029,
      "meanColdCallUsd": 0.050225868421052616,
      "meanCachedTurnUsd": 0.050225868421052616,
      "coldUsdPerCorrect": 0.05142888323353292,
      "cachedUsdPerCorrect": 0.05142888323353292
    },
    {
      "condition": "JSX \u002B shadcn/ui",
      "conditionId": "jsx_shadcn",
      "provider": "OpenAI GPT",
      "providerModelId": "gpt-5.6-terra@low",
      "runs": 168,
      "passRate": 0.9821428571428571,
      "meanColdCallUsd": 0.023596214285714283,
      "meanCachedTurnUsd": 0.023596214285714283,
      "coldUsdPerCorrect": 0.024025236363636363,
      "cachedUsdPerCorrect": 0.024025236363636363
    },
    {
      "condition": "JSX \u002B shadcn/ui",
      "conditionId": "jsx_shadcn",
      "provider": "xAI Grok",
      "providerModelId": "grok-4.5",
      "runs": 183,
      "passRate": 0.994535519125683,
      "meanColdCallUsd": 0.013729066666666663,
      "meanCachedTurnUsd": 0.013729066666666663,
      "coldUsdPerCorrect": 0.013804501098901094,
      "cachedUsdPerCorrect": 0.013804501098901094
    },
    {
      "condition": "Feliz (bare)",
      "conditionId": "feliz_bare",
      "runs": 1200,
      "passRate": 0.7983333333333333,
      "meanColdCallUsd": 0.05619536487500001,
      "meanCachedTurnUsd": 0.056102722791666684,
      "coldUsdPerCorrect": 0.07039085370563675,
      "cachedUsdPerCorrect": 0.07027480934237998
    },
    {
      "condition": "Feliz (bare)",
      "conditionId": "feliz_bare",
      "provider": "Claude (Opus)",
      "providerModelId": "claude-fable-5@low",
      "runs": 171,
      "passRate": 0.8421052631578947,
      "meanColdCallUsd": 0.09900073099415202,
      "meanCachedTurnUsd": 0.09881498245614034,
      "coldUsdPerCorrect": 0.11756336805555553,
      "cachedUsdPerCorrect": 0.11734279166666667
    },
    {
      "condition": "Feliz (bare)",
      "conditionId": "feliz_bare",
      "provider": "Claude (Opus)",
      "providerModelId": "claude-opus-4-8@low",
      "runs": 171,
      "passRate": 0.7777777777777778,
      "meanColdCallUsd": 0.04904159941520465,
      "meanCachedTurnUsd": 0.04857722807017541,
      "coldUsdPerCorrect": 0.06305348496240597,
      "cachedUsdPerCorrect": 0.062456436090225526
    },
    {
      "condition": "Feliz (bare)",
      "conditionId": "feliz_bare",
      "provider": "Google Gemini",
      "providerModelId": "gemini-3.1-pro-preview",
      "runs": 171,
      "passRate": 0.847953216374269,
      "meanColdCallUsd": 0.0874244912280702,
      "meanCachedTurnUsd": 0.0874244912280702,
      "coldUsdPerCorrect": 0.10310060689655175,
      "cachedUsdPerCorrect": 0.10310060689655175
    },
    {
      "condition": "Feliz (bare)",
      "conditionId": "feliz_bare",
      "provider": "Google Gemini",
      "providerModelId": "gemini-3.6-flash",
      "runs": 165,
      "passRate": 0.896969696969697,
      "meanColdCallUsd": 0.05856235454545452,
      "meanCachedTurnUsd": 0.05856235454545452,
      "coldUsdPerCorrect": 0.06528911148648646,
      "cachedUsdPerCorrect": 0.06528911148648646
    },
    {
      "condition": "Feliz (bare)",
      "conditionId": "feliz_bare",
      "provider": "OpenAI GPT",
      "providerModelId": "gpt-5.6-sol@low",
      "runs": 171,
      "passRate": 0.8421052631578947,
      "meanColdCallUsd": 0.05400191228070176,
      "meanCachedTurnUsd": 0.05400191228070176,
      "coldUsdPerCorrect": 0.06412727083333335,
      "cachedUsdPerCorrect": 0.06412727083333335
    },
    {
      "condition": "Feliz (bare)",
      "conditionId": "feliz_bare",
      "provider": "OpenAI GPT",
      "providerModelId": "gpt-5.6-terra@low",
      "runs": 168,
      "passRate": 0.7261904761904762,
      "meanColdCallUsd": 0.03187844196428572,
      "meanCachedTurnUsd": 0.03187844196428572,
      "coldUsdPerCorrect": 0.04389818237704919,
      "cachedUsdPerCorrect": 0.04389818237704919
    },
    {
      "condition": "Feliz (bare)",
      "conditionId": "feliz_bare",
      "provider": "xAI Grok",
      "providerModelId": "grok-4.5",
      "runs": 183,
      "passRate": 0.6666666666666666,
      "meanColdCallUsd": 0.01593944043715847,
      "meanCachedTurnUsd": 0.01593944043715847,
      "coldUsdPerCorrect": 0.023909160655737702,
      "cachedUsdPerCorrect": 0.023909160655737702
    },
    {
      "condition": "Dash (Python)",
      "conditionId": "dash_py",
      "runs": 423,
      "passRate": 0.9905437352245863,
      "meanColdCallUsd": 0.06475062529550826,
      "meanCachedTurnUsd": 0.0645662990543735,
      "coldUsdPerCorrect": 0.06536876968973745,
      "cachedUsdPerCorrect": 0.06518268377088303
    },
    {
      "condition": "Dash (Python)",
      "conditionId": "dash_py",
      "provider": "Claude (Opus)",
      "providerModelId": "claude-fable-5@low",
      "runs": 60,
      "passRate": 1,
      "meanColdCallUsd": 0.09870316666666666,
      "meanCachedTurnUsd": 0.09783683333333333,
      "coldUsdPerCorrect": 0.09870316666666666,
      "cachedUsdPerCorrect": 0.09783683333333333
    },
    {
      "condition": "Dash (Python)",
      "conditionId": "dash_py",
      "provider": "Claude (Opus)",
      "providerModelId": "claude-opus-4-8@low",
      "runs": 60,
      "passRate": 0.95,
      "meanColdCallUsd": 0.05048033333333333,
      "meanCachedTurnUsd": 0.05004716666666666,
      "coldUsdPerCorrect": 0.05313719298245614,
      "cachedUsdPerCorrect": 0.05268122807017544
    },
    {
      "condition": "Dash (Python)",
      "conditionId": "dash_py",
      "provider": "Google Gemini",
      "providerModelId": "gemini-3.1-pro-preview",
      "runs": 60,
      "passRate": 0.9833333333333333,
      "meanColdCallUsd": 0.07782620000000003,
      "meanCachedTurnUsd": 0.07782620000000003,
      "coldUsdPerCorrect": 0.07914528813559325,
      "cachedUsdPerCorrect": 0.07914528813559325
    },
    {
      "condition": "Dash (Python)",
      "conditionId": "dash_py",
      "provider": "Google Gemini",
      "providerModelId": "gemini-3.6-flash",
      "runs": 60,
      "passRate": 1,
      "meanColdCallUsd": 0.09138502499999998,
      "meanCachedTurnUsd": 0.09138502499999998,
      "coldUsdPerCorrect": 0.09138502499999998,
      "cachedUsdPerCorrect": 0.09138502499999998
    },
    {
      "condition": "Dash (Python)",
      "conditionId": "dash_py",
      "provider": "OpenAI GPT",
      "providerModelId": "gpt-5.6-sol@low",
      "runs": 60,
      "passRate": 1,
      "meanColdCallUsd": 0.07708537499999998,
      "meanCachedTurnUsd": 0.07708537499999998,
      "coldUsdPerCorrect": 0.07708537499999998,
      "cachedUsdPerCorrect": 0.07708537499999998
    },
    {
      "condition": "Dash (Python)",
      "conditionId": "dash_py",
      "provider": "OpenAI GPT",
      "providerModelId": "gpt-5.6-terra@low",
      "runs": 60,
      "passRate": 1,
      "meanColdCallUsd": 0.03780327499999999,
      "meanCachedTurnUsd": 0.03780327499999999,
      "coldUsdPerCorrect": 0.03780327499999999,
      "cachedUsdPerCorrect": 0.03780327499999999
    },
    {
      "condition": "Dash (Python)",
      "conditionId": "dash_py",
      "provider": "xAI Grok",
      "providerModelId": "grok-4.5",
      "runs": 63,
      "passRate": 1,
      "meanColdCallUsd": 0.02210336507936508,
      "meanCachedTurnUsd": 0.02210336507936508,
      "coldUsdPerCorrect": 0.02210336507936508,
      "cachedUsdPerCorrect": 0.02210336507936508
    }
  ],
  "costPerCorrectNote": "Expected spend per correct artifact: mean per-call USD over ALL primary cells (failures included; repair cells excluded) / pass rate. Each condition passes its OWN gate - fuaran: strict wire parse AND judge; code baselines: a judged render gate over code never compiled or executed in-harness - so cross-condition comparison is conservative toward the baselines.",
  "costPerCorrectCaveat": "spend totals undercount whenever attempts were abandoned client-side (timeouts bill the provider but return no usage payload) \u2014 see the harness cost-undercount caveat",
  "sessionEconomics": {
    "cohort": "618-stage2-20260807",
    "excludedTasks": [
      {
        "id": "tier-b-001",
        "reason": "pre-v3 probe scaffold: identity figures structurally unfair (authoring protocol v3 note)"
      }
    ],
    "arms": [
      {
        "condition": "feliz_bare",
        "modelPin": "claude-opus-4-8",
        "cells": 24,
        "usdAtTurn3": 0.16512339583333333,
        "cellsAtTurn3": 24,
        "usdAtTurn5": 0.5593513333333334,
        "cellsAtTurn5": 6,
        "identityMean": 1,
        "identityCells": 24,
        "pricingRetrieved": "2026-07-30"
      },
      {
        "condition": "feliz_bare",
        "modelPin": "claude-opus-4-8@low",
        "cells": 24,
        "usdAtTurn3": 0.13218589583333334,
        "cellsAtTurn3": 24,
        "usdAtTurn5": 0.42787908333333335,
        "cellsAtTurn5": 6,
        "identityMean": 1,
        "identityCells": 24,
        "pricingRetrieved": "2026-07-30"
      },
      {
        "condition": "feliz_bare",
        "modelPin": "gemini-3.1-pro-preview",
        "cells": 24,
        "usdAtTurn3": 0.14046783333333335,
        "cellsAtTurn3": 24,
        "usdAtTurn5": 0.3589599666666667,
        "cellsAtTurn5": 6,
        "identityMean": 1,
        "identityCells": 24,
        "pricingRetrieved": "2026-07-30"
      },
      {
        "condition": "feliz_bare",
        "modelPin": "gpt-4o",
        "cells": 24,
        "usdAtTurn3": 0.02625416666666667,
        "cellsAtTurn3": 24,
        "usdAtTurn5": 0.07226791666666667,
        "cellsAtTurn5": 6,
        "identityMean": 1,
        "identityCells": 24,
        "pricingRetrieved": "2026-07-30"
      },
      {
        "condition": "fuaran",
        "modelPin": "claude-opus-4-8",
        "cells": 24,
        "usdAtTurn3": 0.19215895833333335,
        "cellsAtTurn3": 24,
        "usdAtTurn5": 0.34094083333333336,
        "cellsAtTurn5": 6,
        "identityMean": 0.9861111111111112,
        "identityCells": 24,
        "pricingRetrieved": "2026-07-30"
      },
      {
        "condition": "fuaran",
        "modelPin": "claude-opus-4-8@low",
        "cells": 24,
        "usdAtTurn3": 0.15023958333333334,
        "cellsAtTurn3": 24,
        "usdAtTurn5": 0.3338925,
        "cellsAtTurn5": 6,
        "identityMean": 0.986111111111111,
        "identityCells": 24,
        "pricingRetrieved": "2026-07-30"
      },
      {
        "condition": "fuaran",
        "modelPin": "gemini-3.1-pro-preview",
        "cells": 24,
        "usdAtTurn3": 0.19021734166666668,
        "cellsAtTurn3": 24,
        "usdAtTurn5": 0.3297098666666667,
        "cellsAtTurn5": 6,
        "identityMean": 1,
        "identityCells": 24,
        "pricingRetrieved": "2026-07-30"
      },
      {
        "condition": "fuaran",
        "modelPin": "gpt-4o",
        "cells": 24,
        "usdAtTurn3": 0.0935509375,
        "cellsAtTurn3": 24,
        "usdAtTurn5": 0.17769875,
        "cellsAtTurn5": 6,
        "identityMean": 0.9583333333333335,
        "identityCells": 24,
        "pricingRetrieved": "2026-07-30"
      },
      {
        "condition": "jsx_shadcn",
        "modelPin": "claude-opus-4-8",
        "cells": 24,
        "usdAtTurn3": 0.17562533333333336,
        "cellsAtTurn3": 24,
        "usdAtTurn5": 0.5451363333333333,
        "cellsAtTurn5": 6,
        "identityMean": 1,
        "identityCells": 24,
        "pricingRetrieved": "2026-07-30"
      },
      {
        "condition": "jsx_shadcn",
        "modelPin": "claude-opus-4-8@low",
        "cells": 24,
        "usdAtTurn3": 0.14159591666666668,
        "cellsAtTurn3": 24,
        "usdAtTurn5": 0.443808,
        "cellsAtTurn5": 6,
        "identityMean": 1,
        "identityCells": 24,
        "pricingRetrieved": "2026-07-30"
      },
      {
        "condition": "jsx_shadcn",
        "modelPin": "gemini-3.1-pro-preview",
        "cells": 25,
        "usdAtTurn3": 0.12795984,
        "cellsAtTurn3": 25,
        "usdAtTurn5": 0.2658003333333333,
        "cellsAtTurn5": 6,
        "identityMean": 1,
        "identityCells": 25,
        "pricingRetrieved": "2026-07-30"
      },
      {
        "condition": "jsx_shadcn",
        "modelPin": "gpt-4o",
        "cells": 24,
        "usdAtTurn3": 0.024686770833333333,
        "cellsAtTurn3": 24,
        "usdAtTurn5": 0.0650575,
        "cellsAtTurn5": 6,
        "identityMean": 1,
        "identityCells": 24,
        "pricingRetrieved": "2026-07-30"
      }
    ],
    "note": "Tier-B multi-turn session economics: cumulative billed USD through turn k, mean over sessions that reached turn k (denominator beside each figure); identity = fraction of prompt-named element ids present in both the first and final emission, mean over measured cells. Multi-turn cells publish here only \u2014 parsePassRate remains a single-shot claim."
  }
}
