{ "method": { "version": "2.0.0", "aggregation": "Equal mean of supported private-evaluation shortfalls", "minPrivateEvaluations": 2, "minPublicPredictors": 2, "minReferenceModels": 8, "minReferenceProviders": 3, "minCorrelation": 0.35 }, "retrievedAt": "2026-10-13", "source": "https://artificialanalysis.ai/evaluations/frontiercode", "coverage": { "models": 22, "providers": 14, "benchmarkCells": 396, "observedCells": 339, "scored": 22 }, "comparisons": [ { "key": "frontierCode", "label": "Coding", "predictors": [ "scicode", "terminalBench40" ], "scored": 22, "correlations": [ { "key": "scicode", "rho": 0.6273291925465838, "models": 22, "slope": 0.8969070422535184, "intercept": -8.353182042253401 }, { "key": "terminalBench40", "rho": 0.8239618671108515, "models": 22, "slope": 0.3503518931159409, "intercept": 29.48713808373592 } ] }, { "key": "briefcaseRubric", "label": "Knowledge work", "predictors": [ "gdpval", "gdpPdf" ], "scored": 22, "correlations": [ { "key": "gdpval", "rho": 0.8893280632411067, "models": 22, "slope": 0.05020313412369358, "intercept": -29.35228004390013 }, { "key": "gdpPdf", "rho": 0.7275301427546231, "models": 22, "slope": 1.0468848905593426, "intercept": 30.693878598794083 } ] }, { "key": "automation", "label": "Workflows", "predictors": [ "enterpriseOps", "terminalBench40" ], "scored": 0, "correlations": [ { "key": "enterpriseOps", "rho": 0.13333333333333333, "models": 9, "slope": 1.4449195397141994, "intercept": -15.659996122486312 }, { "key": "terminalBench40", "rho": 0.7403223908883507, "models": 22, "slope": 0.5257650327737762, "intercept": 38.25812805263483 } ] }, { "key": "analyst", "label": "Data analysis", "predictors": [ "scicode", "gdpPdf" ], "scored": 16, "correlations": [ { "key": "scicode", "rho": 0.6833586203385283, "models": 16, "slope": 1.5498069498069493, "intercept": -48.252895752895725 }, { "key": "gdpPdf", "rho": 0.7505537099314585, "models": 16, "slope": 1.8867924528301887, "intercept": 0.9316037735849054 } ] }, { "key": "omniscienceAccuracy", "label": "Knowledge", "predictors": [ "gpqa", "hle" ], "scored": 13, "correlations": [ { "key": "gpqa", "rho": 0.7290240734419666, "models": 13, "slope": 4.563607594936702, "intercept": -380.09514767932444 }, { "key": "hle", "rho": 0.7662337662337663, "models": 22, "slope": 1.4642167577413376, "intercept": -25.069216757740776 } ] }, { "key": "harveyLab", "label": "Legal work", "predictors": [ "gdpPdf", "lcr" ], "scored": 1, "correlations": [ { "key": "gdpPdf", "rho": 0.7184037438917437, "models": 16, "slope": 0.38156288156288143, "intercept": -4.439865689865687 }, { "key": "lcr", "rho": 0.22057741053654933, "models": 16, "slope": 0.39062499999998734, "intercept": -29.435763888887816 } ] }, { "key": "itbench", "label": "Operations", "predictors": [ "terminalBench40", "terminalBench21" ], "scored": 0, "correlations": [ { "key": "terminalBench40", "rho": -0.2556921609434964, "models": 12, "slope": -0.09776767259198209, "intercept": 49.85720683478027 }, { "key": "terminalBench21", "rho": 0.21557272714438663, "models": 8, "slope": 0.3142655367231647, "intercept": 20.809792843691053 } ] }, { "key": "mlcr", "label": "Medical reasoning", "predictors": [ "lcr", "gdpPdf" ], "scored": 0, "correlations": [ { "key": "lcr", "rho": 0.1181641626690443, "models": 21, "slope": 0.4040404040404091, "intercept": -11.161616161616564 }, { "key": "gdpPdf", "rho": 0.437540756236812, "models": 21, "slope": 1.109321882268741, "intercept": 6.350867775988538 } ] } ], "models": [ { "id": "minimax-m3", "name": "MiniMax-M3", "shortName": "MiniMax-M3", "provider": "MiniMax", "release": "2026-06-01", "scores": { "frontierCode": 23.0037666666667, "briefcaseRubric": 30.303030303030305, "automation": 21.25125760074941, "analyst": 10, "omniscienceAccuracy": 16.7, "harveyLab": 0, "itbench": null, "mlcr": 17.2222222222222, "scicode": 47.1064814814815, "terminalBench40": 2.02020202020202, "terminalBench21": 65.1685393258427, "terminalBenchScience": 0.476190476190476, "gdpval": 1242.26, "gdpPdf": 9.8, "enterpriseOps": 32.10981796478663, "gpqa": 92.9292929292929, "hle": 38.9712696941613, "lcr": 83 }, "status": "scored", "index": 42.14430745170428, "discordance": 13.255622184594143, "privateEvaluations": 4, "extrapolated": true, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 23.0037666666667, "expected": 32.620660214534155, "gapPoints": 9.616893547867456, "index": 29.480989914430495, "discordance": 6.40760410780712, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.5805194805194805, "weight": 0.33700286726260753, "slope": 0.8379024189723281, "intercept": -4.850316996047233, "expected": 34.620317786561294, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 21, "discordance": 9.523809523809524, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.7988304089244538, "weight": 0.63813002222241, "slope": 0.3144044444444429, "intercept": 30.929460493827214, "expected": 31.564620987654372, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 21, "discordance": 4.761904761904762, "extrapolated": false } ], "sensitivityRange": { "min": 25.16427925482481, "max": 32.387584276544956, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 30.303030303030305, "expected": 37.0409001187182, "gapPoints": 6.737869815687894, "index": 18.19035119042096, "discordance": 1.8566098107028215, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.8727272727272727, "weight": 0.7616528925619834, "slope": 0.04860899764344703, "intercept": -26.70964856532425, "expected": 33.67536484722426, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.697659887666064, "weight": 0.48672931885822507, "slope": 0.9622846488277563, "intercept": 32.87703052363457, "expected": 42.307420082146585, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 21, "discordance": 4.761904761904762, "extrapolated": false } ], "sensitivityRange": { "min": 16.932545593150913, "max": 19.908122702592394, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7039327362613372, "weight": 0.49552129718037335, "slope": 0.45054755328899165, "intercept": 42.83518608908643, "expected": 43.745383166437925, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 21, "discordance": 4.761904761904762, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 10, "expected": 25.92562732511409, "gapPoints": 15.925627325114089, "index": 61.42812717857352, "discordance": 9.101943930120251, "extrapolated": true, "predictors": [ { "key": "scicode", "rho": 0.6368525395992862, "weight": 0.4055811571940604, "slope": 1.324528301886794, "intercept": -32.3584905660379, "expected": 30.0353773584905, "referenceModels": 15, "referenceProviders": 8, "discordantPairs": 3, "comparablePairs": 15, "discordance": 20, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.6968612668106766, "weight": 0.485615625180981, "slope": 1.596467391304348, "intercept": 6.847826086956523, "expected": 22.493206521739133, "referenceModels": 15, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 15, "discordance": 0, "extrapolated": true } ], "sensitivityRange": { "min": 53.21212752998193, "max": 64.66748842688521, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 16.7, "expected": 41.21193850048623, "gapPoints": 24.511938500486227, "index": 59.47776152339213, "discordance": 35.65633088974638, "extrapolated": false, "predictors": [ { "key": "gpqa", "rho": 0.8231186002975568, "weight": 0.6775242301558091, "slope": 4.694516129032306, "intercept": -392.04220430107983, "expected": 44.21586021505351, "referenceModels": 12, "referenceProviders": 9, "discordantPairs": 6, "comparablePairs": 12, "discordance": 50, "extrapolated": false }, { "key": "hle", "rho": 0.7649350649350649, "weight": 0.585125653567212, "slope": 1.2430172041606822, "intercept": -10.708298298298288, "expected": 37.73366039953002, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 4, "comparablePairs": 21, "discordance": 19.047619047619047, "extrapolated": false } ], "sensitivityRange": { "min": 55.095438648291484, "max": 61.51338734567329, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6729401487337795, "weight": 0.4528484437778413, "slope": 0.38665038665038665, "intercept": -4.94912494912495, "expected": 0, "referenceModels": 15, "referenceProviders": 9, "discordantPairs": 0, "comparablePairs": 13, "discordance": 0, "extrapolated": true } ] }, { "key": "itbench", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.3952525613618202, "weight": 0.15622458726307945, "slope": 1.109321882268741, "intercept": 5.915190552871757, "expected": 16.78654499910542, "referenceModels": 20, "referenceProviders": 12, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": true } ] } ], "sensitivityRange": { "min": 38.03401258098085, "max": 42.872103962873716, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "muse-glimmer", "name": "Muse Glimmer (High)", "shortName": "Muse Glimmer (high)", "provider": "Meta", "release": "2026-08-10", "scores": { "frontierCode": 13.917833333333299, "briefcaseRubric": 11.514522821576763, "automation": 6.831776269647931, "analyst": 20, "omniscienceAccuracy": 26.98333333333333, "harveyLab": null, "itbench": null, "mlcr": 20, "scicode": 44.907407407407405, "terminalBench40": 0.505050505050505, "terminalBench21": 51.685393258427005, "terminalBenchScience": 0, "gdpval": 786.36, "gdpPdf": 10, "enterpriseOps": 34.73589973142345, "gpqa": 83.5353535353535, "hle": 21.9647822057461, "lcr": 83.3333333333333 }, "status": "scored", "index": 27.57284533846618, "discordance": 1.3966474141263676, "privateEvaluations": 4, "extrapolated": true, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 13.917833333333299, "expected": 31.996185513336357, "gapPoints": 18.07835218000306, "index": 56.50158570454533, "discordance": 1.7834150023941708, "extrapolated": true, "predictors": [ { "key": "scicode", "rho": 0.6090225563909775, "weight": 0.3709084741930013, "slope": 0.8379024189723281, "intercept": -4.554427075098587, "expected": 33.07359822134392, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 20, "discordance": 5, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8179084331947194, "weight": 0.6689742050910408, "slope": 0.31348149396135105, "intercept": 31.240496933709128, "expected": 31.398820920558297, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": true } ], "sensitivityRange": { "min": 53.44941070297256, "max": 58.10695844186457, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 11.514522821576763, "expected": 22.250646959077322, "gapPoints": 10.736124137500559, "index": 48.250840333973635, "discordance": 3.8031746541113, "extrapolated": true, "predictors": [ { "key": "gdpval", "rho": 0.8631578947368421, "weight": 0.7450415512465375, "slope": 0.0498177236674861, "intercept": -29.2215242185339, "expected": 9.953140964630471, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": true }, { "key": "gdpPdf", "rho": 0.6762055861313887, "weight": 0.45725399471529493, "slope": 0.9515835295891515, "intercept": 32.772149878833346, "expected": 42.28798517472486, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 20, "discordance": 10, "extrapolated": false } ], "sensitivityRange": { "min": 44.325039755992194, "max": 52.15277871911041, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7253576169270557, "weight": 0.5261436724340973, "slope": 0.482486242343317, "intercept": 40.924800998349994, "expected": 41.16848091872541, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": true } ] }, { "key": "analyst", "status": "scored", "observed": 20, "expected": 21.172749112364116, "gapPoints": 1.1727491123641158, "index": 5.538955315345767, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "scicode", "rho": 0.68646906226167, "weight": 0.47123977344241663, "slope": 1.5567567567567595, "intercept": -44.594594594594774, "expected": 25.315315315315253, "referenceModels": 14, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 14, "discordance": 0, "extrapolated": true }, { "key": "gdpPdf", "rho": 0.8156752766113337, "weight": 0.6653261568749756, "slope": 2.0643939393939386, "intercept": -2.405303030303017, "expected": 18.238636363636367, "referenceModels": 14, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 14, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 18.514847019770173, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 26.98333333333333, "expected": 1.6610480981154314, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "gpqa", "rho": 0.6636363636363637, "weight": 0.4404132231404959, "slope": 5.869285714285595, "intercept": -504.46428571427464, "expected": 0, "referenceModels": 11, "referenceProviders": 9, "discordantPairs": 0, "comparablePairs": 11, "discordance": 0, "extrapolated": true }, { "key": "hle", "rho": 0.7609022556390977, "weight": 0.5789722426366669, "slope": 1.5946485152621062, "intercept": -32.1015313951557, "expected": 2.9245759372928433, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": true } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 11, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "itbench", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.4210992672281177, "weight": 0.17732459286005767, "slope": 1.1316872427983518, "intercept": 2.880658436214059, "expected": 14.197530864197576, "referenceModels": 19, "referenceProviders": 12, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": false } ] } ], "sensitivityRange": { "min": 24.968595909186618, "max": 31.84153202770583, "validRuns": 11, "attemptedRuns": 13 } }, { "id": "inkling", "name": "Inkling (Xhigh)", "shortName": "Inkling (xhigh)", "provider": "Thinking Machines", "release": "2026-07-15", "scores": { "frontierCode": 15.8388333333333, "briefcaseRubric": 19.29292929292929, "automation": 4.984889684475201, "analyst": 23.75, "omniscienceAccuracy": 41.55, "harveyLab": 0, "itbench": null, "mlcr": 12.2222222222222, "scicode": 46.9907407407407, "terminalBench40": 1.01010101010101, "terminalBench21": 55.0561797752809, "terminalBenchScience": 0, "gdpval": 1075.6, "gdpPdf": 12.8, "enterpriseOps": 37.95881826320501, "gpqa": 87.1717171717172, "hle": 31.881371640407803, "lcr": 77.3333333333333 }, "status": "scored", "index": 26.03363699260306, "discordance": 2.3191781309079262, "privateEvaluations": 4, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 15.8388333333333, "expected": 32.437976216168465, "gapPoints": 16.599142882835167, "index": 51.17194356459713, "discordance": 3.2984181837680806, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.5805194805194805, "weight": 0.33700286726260753, "slope": 0.8379024189723281, "intercept": -4.850316996047233, "expected": 34.523338339920926, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 21, "discordance": 9.523809523809524, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.797530440805781, "weight": 0.6360548040118633, "slope": 0.3121329444444436, "intercept": 31.017796604938297, "expected": 31.33308240740743, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 47.79349079000013, "max": 53.17323187628248, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 19.29292929292929, "expected": 33.42037539255978, "gapPoints": 14.127446099630486, "index": 42.27195515815664, "discordance": 5.9782943398636235, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.8727272727272727, "weight": 0.7616528925619834, "slope": 0.048029865790845824, "intercept": -26.042601901991198, "expected": 25.618321742642564, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7104325768547014, "weight": 0.5047144462564113, "slope": 0.9577583308445473, "intercept": 32.93496739381965, "expected": 45.194274028629856, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 20, "discordance": 15, "extrapolated": false } ], "sensitivityRange": { "min": 40.70328823466174, "max": 43.87959162834417, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7026327681426644, "weight": 0.4936928068678232, "slope": 0.45054755328899165, "intercept": 42.83518608908643, "expected": 43.290284627762176, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 21, "discordance": 4.761904761904762, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 23.75, "expected": 26.59296008752736, "gapPoints": 2.8429600875273593, "index": 10.690649247658465, "discordance": 0, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6261190698307589, "weight": 0.3920250896057348, "slope": 1.4025974025973915, "intercept": -37.581168831168156, "expected": 28.327922077922167, "referenceModels": 15, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 15, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7076236029776369, "weight": 0.5007311634910523, "slope": 1.869517543859649, "intercept": 1.3048245614035103, "expected": 25.234649122807017, "referenceModels": 15, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 15, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 21.631264931802065, "validRuns": 12, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 41.55, "expected": 10.94580394747126, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": false, "predictors": [ { "key": "gpqa", "rho": 0.8231186002975568, "weight": 0.6775242301558091, "slope": 6.4155882352938285, "intercept": -555.5352941176201, "expected": 3.722549019609801, "referenceModels": 12, "referenceProviders": 9, "discordantPairs": 0, "comparablePairs": 12, "discordance": 0, "extrapolated": false }, { "key": "hle", "rho": 0.8181818181818182, "weight": 0.6694214876033059, "slope": 1.5946485152621062, "intercept": -32.58309209798696, "expected": 18.256489852908757, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6857657179657172, "weight": 0.47027461993703557, "slope": 0.38956639566395657, "intercept": -4.085365853658536, "expected": 0.9010840108401084, "referenceModels": 15, "referenceProviders": 9, "discordantPairs": 1, "comparablePairs": 12, "discordance": 8.333333333333334, "extrapolated": false } ] }, { "key": "itbench", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.3943503104578523, "weight": 0.1555121673582045, "slope": 0.9920634920634998, "intercept": 8.789682539682453, "expected": 21.488095238095248, "referenceModels": 20, "referenceProviders": 12, "discordantPairs": 3, "comparablePairs": 19, "discordance": 15.789473684210526, "extrapolated": false } ] } ], "sensitivityRange": { "min": 22.918270604586077, "max": 29.47163655785871, "validRuns": 12, "attemptedRuns": 13 } }, { "id": "qwen3-8-27b", "name": "Qwen3.8 27B (Xhigh)", "shortName": "Qwen3.8 27B (xhigh)", "provider": "Alibaba", "release": "2026-08-14", "scores": { "frontierCode": 34.2324, "briefcaseRubric": 47.80388151174668, "automation": 48.24244368224193, "analyst": 18.75, "omniscienceAccuracy": 15.583333333333332, "harveyLab": null, "itbench": null, "mlcr": 21.6666666666667, "scicode": 46.6435185185185, "terminalBench40": 5.55555555555556, "terminalBench21": 79.7752808988764, "terminalBenchScience": 0.476190476190476, "gdpval": 1419.13, "gdpPdf": 16.6, "enterpriseOps": 44.225604297224706, "gpqa": 90.5050505050505, "hle": 33.920296570899, "lcr": 82 }, "status": "scored", "index": 22.524199895205385, "discordance": 9.45901468609461, "privateEvaluations": 4, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 34.2324, "expected": 30.106259724907456, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6090225563909775, "weight": 0.3709084741930013, "slope": 1.1394282115384504, "intercept": -24.00258573717881, "expected": 29.144355148237313, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8084310760975514, "weight": 0.6535608048002449, "slope": 0.3704444754098352, "intercept": 28.59413442622951, "expected": 30.652159289617483, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 47.80388151174668, "expected": 44.02496006771024, "gapPoints": 0, "index": 0, "discordance": 1.9029326251895669, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.9007518796992481, "weight": 0.8113539487817287, "slope": 0.05115421850076689, "intercept": -31.304852364651993, "expected": 41.28963373634133, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7060599156233736, "weight": 0.4985206044500854, "slope": 0.9898989898989903, "intercept": 32.04444444444444, "expected": 48.47676767676768, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 20, "discordance": 5, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7369218095898536, "weight": 0.5430537534491845, "slope": 0.5548092271266981, "intercept": 36.46853627937398, "expected": 39.550809763411195, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 18.75, "expected": 29.07346289386352, "gapPoints": 10.323462893863521, "index": 35.50819842669128, "discordance": 14.724214920609636, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6674025000565448, "weight": 0.44542609708172626, "slope": 1.3846153846153813, "intercept": -38.846153846153655, "expected": 25.73717948717949, "referenceModels": 14, "referenceProviders": 8, "discordantPairs": 1, "comparablePairs": 14, "discordance": 7.142857142857143, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7097130242825607, "weight": 0.5036925768362985, "slope": 1.6369047619047619, "intercept": 4.851190476190476, "expected": 32.023809523809526, "referenceModels": 14, "referenceProviders": 8, "discordantPairs": 3, "comparablePairs": 14, "discordance": 21.428571428571427, "extrapolated": false } ], "sensitivityRange": { "min": 29.305475486498448, "max": 43.37820728901375, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 15.583333333333332, "expected": 34.31590686343869, "gapPoints": 18.73257353010536, "index": 54.58860115413027, "discordance": 21.208911198579237, "extrapolated": false, "predictors": [ { "key": "gpqa", "rho": 0.7608220194556027, "weight": 0.5788501452885014, "slope": 4.070991461100498, "intercept": -333.1982922201071, "expected": 35.246995572422804, "referenceModels": 11, "referenceProviders": 9, "discordantPairs": 3, "comparablePairs": 11, "discordance": 27.272727272727273, "extrapolated": false }, { "key": "hle", "rho": 0.7518796992481203, "weight": 0.5653230821414438, "slope": 1.16499634735022, "intercept": -6.1544825627922535, "expected": 33.36253904334127, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 20, "discordance": 15, "extrapolated": false } ], "sensitivityRange": { "min": 43.76207602439612, "max": 57.00071649263946, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "itbench", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.43917473339246044, "weight": 0.19287444645033872, "slope": 1.2222222222222232, "intercept": 7.133333333333342, "expected": 27.422222222222246, "referenceModels": 19, "referenceProviders": 12, "discordantPairs": 3, "comparablePairs": 18, "discordance": 16.666666666666668, "extrapolated": false } ] } ], "sensitivityRange": { "min": 20.0321928595077, "max": 24.81442818067815, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "glm-5-3-flash", "name": "GLM-5.3-Flash", "shortName": "GLM-5.3-Flash", "provider": "Z AI", "release": "2026-08-26", "scores": { "frontierCode": 40.9957333333333, "briefcaseRubric": 46.35897435897436, "automation": 60.36862782167782, "analyst": null, "omniscienceAccuracy": 27.500000000000004, "harveyLab": 0.8333333333333334, "itbench": 51.24293785310739, "mlcr": 51.11111111111109, "scicode": 51.6203703703704, "terminalBench40": 32.8282828282828, "terminalBench21": 84.2696629213483, "terminalBenchScience": 4.76190476190476, "gdpval": 1644.13, "gdpPdf": 15.4, "enterpriseOps": 33.21396598030439, "gpqa": 91.2121212121212, "hle": 39.8517145505097, "lcr": 80 }, "status": "scored", "index": 13.069931912862303, "discordance": 14.145898841541275, "privateEvaluations": 3, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 40.9957333333333, "expected": 39.91842254955724, "gapPoints": 0, "index": 0, "discordance": 10.10029961974839, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6556390977443609, "weight": 0.4298626264910396, "slope": 0.9029115211267607, "intercept": -8.737149007237932, "expected": 37.871478124999975, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8735894120874602, "weight": 0.7631584609113143, "slope": 0.371674437704917, "intercept": 28.869966976320608, "expected": 41.0714005373406, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 19, "discordance": 15.789473684210526, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 46.35897435897436, "expected": 51.302325688569546, "gapPoints": 4.943351329595188, "index": 9.63572559966145, "discordance": 20.19677061737886, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.9187969924812031, "weight": 0.8441879133925039, "slope": 0.05115421850076689, "intercept": -29.70392194187964, "expected": 54.40026332178623, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 6, "comparablePairs": 20, "discordance": 30, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7379526441077516, "weight": 0.544574104945622, "slope": 1.087969439806522, "intercept": 29.745235199486807, "expected": 46.499964572507245, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 20, "discordance": 5, "extrapolated": false } ], "sensitivityRange": { "min": 7.540713346433849, "max": 10.75330768371253, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7599701173198405, "weight": 0.5775545792191322, "slope": 0.643539551423367, "intercept": 31.17516208959541, "expected": 52.30146049490795, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 19, "discordance": 10.526315789473685, "extrapolated": false } ] }, { "key": "analyst", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 27.500000000000004, "expected": 39.04811772346887, "gapPoints": 11.548117723468867, "index": 29.57407013892546, "discordance": 12.140626287496573, "extrapolated": false, "predictors": [ { "key": "gpqa", "rho": 0.7425987375524745, "weight": 0.5514528850145289, "slope": 4.070991461100498, "intercept": -333.1982922201071, "expected": 38.125474383301935, "referenceModels": 11, "referenceProviders": 9, "discordantPairs": 1, "comparablePairs": 11, "discordance": 9.090909090909092, "extrapolated": false }, { "key": "hle", "rho": 0.7669172932330827, "weight": 0.5881621346599581, "slope": 1.190180743243244, "intercept": -7.51756756756761, "expected": 39.913175675675625, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 20, "discordance": 15, "extrapolated": false } ], "sensitivityRange": { "min": 16.265778837558702, "max": 32.53236734337124, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6578077123308144, "weight": 0.4327109864018994, "slope": 0.39103923647932126, "intercept": -4.430820666902322, "expected": 1.591183574879226, "referenceModels": 14, "referenceProviders": 9, "discordantPairs": 1, "comparablePairs": 14, "discordance": 7.142857142857143, "extrapolated": false } ] }, { "key": "itbench", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6108181781966834, "weight": 0.37309884681551525, "slope": 1.2515262515262529, "intercept": 0.8913308913309024, "expected": 20.164835164835196, "referenceModels": 19, "referenceProviders": 12, "discordantPairs": 0, "comparablePairs": 19, "discordance": 0, "extrapolated": false } ] } ], "sensitivityRange": { "min": 8.553713294023813, "max": 13.870189917108398, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "glm-5-3", "name": "GLM-5.3 (Max)", "shortName": "GLM-5.3 (max)", "provider": "Z AI", "release": "2026-08-18", "scores": { "frontierCode": 40.0752, "briefcaseRubric": 51.34854771784232, "automation": 62.202864996264196, "analyst": null, "omniscienceAccuracy": 33.85, "harveyLab": 0.27777777777777773, "itbench": 46.0828625235405, "mlcr": 48.3333333333333, "scicode": 59.0277777777778, "terminalBench40": 41.9191919191919, "terminalBench21": 83.8951310861423, "terminalBenchScience": 9.52380952380952, "gdpval": 1650.73, "gdpPdf": 11.200000000000001, "enterpriseOps": 36.436884512085946, "gpqa": 91.7171717171717, "hle": 42.2613531047266, "lcr": 79.6666666666667 }, "status": "scored", "index": 9.467257129758595, "discordance": 18.752896219307964, "privateEvaluations": 3, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 40.0752, "expected": 44.4896963881954, "gapPoints": 4.414496388195396, "index": 9.922514079836942, "discordance": 35, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6556390977443609, "weight": 0.4298626264910396, "slope": 0.9029115211267607, "intercept": -8.737149007237932, "expected": 44.55971161482782, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 7, "comparablePairs": 20, "discordance": 35, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8735894120874602, "weight": 0.7631584609113143, "slope": 0.371674437704917, "intercept": 28.869966976320608, "expected": 44.45025906193076, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 7, "comparablePairs": 20, "discordance": 35, "extrapolated": false } ], "sensitivityRange": { "min": 8.457721677005502, "max": 10.480598481148055, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 51.34854771784232, "expected": 49.715730576091296, "gapPoints": 0, "index": 0, "discordance": 9.118062370427317, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.9187969924812031, "weight": 0.8441879133925039, "slope": 0.05115421850076689, "intercept": -29.70392194187964, "expected": 54.73788116389129, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 20, "discordance": 15, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7379526441077516, "weight": 0.544574104945622, "slope": 1.087969439806522, "intercept": 29.745235199486807, "expected": 41.930492925319854, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7599701173198405, "weight": 0.5775545792191322, "slope": 0.643539551423367, "intercept": 31.17516208959541, "expected": 58.1518200533022, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 20, "discordance": 10, "extrapolated": false } ] }, { "key": "analyst", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 33.85, "expected": 41.52317420424987, "gapPoints": 7.673174204249868, "index": 18.479257309438843, "discordance": 12.140626287496573, "extrapolated": false, "predictors": [ { "key": "gpqa", "rho": 0.7425987375524745, "weight": 0.5514528850145289, "slope": 4.070991461100498, "intercept": -333.1982922201071, "expected": 40.18153067678696, "referenceModels": 11, "referenceProviders": 9, "discordantPairs": 1, "comparablePairs": 11, "discordance": 9.090909090909092, "extrapolated": false }, { "key": "hle", "rho": 0.7669172932330827, "weight": 0.5881621346599581, "slope": 1.190180743243244, "intercept": -7.51756756756761, "expected": 42.781081081081076, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 20, "discordance": 15, "extrapolated": false } ], "sensitivityRange": { "min": 6.424759174871314, "max": 21.3868486096703, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6578077123308144, "weight": 0.4327109864018994, "slope": 0.39103923647932126, "intercept": -4.430820666902322, "expected": 0, "referenceModels": 14, "referenceProviders": 9, "discordantPairs": 0, "comparablePairs": 14, "discordance": 0, "extrapolated": false } ] }, { "key": "itbench", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6108181781966834, "weight": 0.37309884681551525, "slope": 1.2515262515262529, "intercept": 0.8913308913309024, "expected": 14.908424908424935, "referenceModels": 19, "referenceProviders": 12, "discordantPairs": 0, "comparablePairs": 19, "discordance": 0, "extrapolated": false } ] } ], "sensitivityRange": { "min": 5.544902480027349, "max": 10.103602483411533, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "muse-spark-1-3", "name": "Muse Spark 1.3 (Max)", "shortName": "Muse Spark 1.3 (max)", "provider": "Meta", "release": "2026-09-02", "scores": { "frontierCode": 40.6458333333333, "briefcaseRubric": 58.686868686868685, "automation": 57.853437865386134, "analyst": 37.5, "omniscienceAccuracy": 43.583333333333336, "harveyLab": 8.888888888888888, "itbench": 33.239171374764595, "mlcr": 43.3333333333333, "scicode": 58.796296296296305, "terminalBench40": 33.3333333333333, "terminalBench21": 84.2696629213483, "terminalBenchScience": 10.952380952380999, "gdpval": 1680.99, "gdpPdf": 26.6, "enterpriseOps": null, "gpqa": 93.53535353535351, "hle": 48.7025023169602, "lcr": 83 }, "status": "scored", "index": 8.41149032797435, "discordance": 24.19613510103104, "privateEvaluations": 4, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 40.6458333333333, "expected": 42.767509206011994, "gapPoints": 2.1216758726786935, "index": 4.960952629853977, "discordance": 29.4132997410004, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6090225563909775, "weight": 0.3709084741930013, "slope": 0.8379024189723281, "intercept": -4.554427075098587, "expected": 44.71113181818182, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 7, "comparablePairs": 20, "discordance": 35, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8179084331947194, "weight": 0.6689742050910408, "slope": 0.31348149396135105, "intercept": 31.240496933709128, "expected": 41.68988006575415, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 5, "comparablePairs": 19, "discordance": 26.31578947368421, "extrapolated": false } ], "sensitivityRange": { "min": 2.9708842774008093, "max": 5.782627941229644, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 58.686868686868685, "expected": 55.876528380628464, "gapPoints": 0, "index": 0, "discordance": 5.704761981166951, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.8631578947368421, "weight": 0.7450415512465375, "slope": 0.0498177236674861, "intercept": -29.2215242185339, "expected": 54.521571089273564, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.6762055861313887, "weight": 0.45725399471529493, "slope": 0.9515835295891515, "intercept": 32.772149878833346, "expected": 58.084271765904774, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 20, "discordance": 15, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7253576169270557, "weight": 0.5261436724340973, "slope": 0.482486242343317, "intercept": 40.924800998349994, "expected": 57.00767574312721, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 5, "comparablePairs": 19, "discordance": 26.31578947368421, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 37.5, "expected": 50.1978931375501, "gapPoints": 12.697893137550103, "index": 25.295669487075653, "discordance": 47.46742783039134, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.68646906226167, "weight": 0.47123977344241663, "slope": 1.5567567567567595, "intercept": -44.594594594594774, "expected": 46.93693693693694, "referenceModels": 14, "referenceProviders": 8, "discordantPairs": 5, "comparablePairs": 13, "discordance": 38.46153846153846, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.8156752766113337, "weight": 0.6653261568749756, "slope": 2.0643939393939386, "intercept": -2.405303030303017, "expected": 52.50757575757575, "referenceModels": 14, "referenceProviders": 8, "discordantPairs": 7, "comparablePairs": 13, "discordance": 53.84615384615385, "extrapolated": false } ], "sensitivityRange": { "min": 18.305117781462116, "max": 27.80882783408575, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 43.583333333333336, "expected": 45.11234367942878, "gapPoints": 1.529010346095447, "index": 3.389339194967774, "discordance": 14.199050851565456, "extrapolated": false, "predictors": [ { "key": "gpqa", "rho": 0.6636363636363637, "weight": 0.4404132231404959, "slope": 5.869285714285595, "intercept": -504.46428571427464, "expected": 44.52142857142826, "referenceModels": 11, "referenceProviders": 9, "discordantPairs": 0, "comparablePairs": 10, "discordance": 0, "extrapolated": false }, { "key": "hle", "rho": 0.7609022556390977, "weight": 0.5789722426366669, "slope": 1.5946485152621062, "intercept": -32.1015313951557, "expected": 45.56184161413417, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 5, "comparablePairs": 20, "discordance": 25, "extrapolated": false } ], "sensitivityRange": { "min": 2.628751303952107, "max": 9.83040066312625, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.7044028000389669, "weight": 0.49618330470273686, "slope": 0.33501683501683505, "intercept": -3.4744107744107753, "expected": 5.437037037037038, "referenceModels": 15, "referenceProviders": 9, "discordantPairs": 1, "comparablePairs": 15, "discordance": 6.666666666666667, "extrapolated": false } ] }, { "key": "itbench", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.4210992672281177, "weight": 0.17732459286005767, "slope": 1.1316872427983518, "intercept": 2.880658436214059, "expected": 32.98353909465022, "referenceModels": 19, "referenceProviders": 12, "discordantPairs": 6, "comparablePairs": 19, "discordance": 31.57894736842105, "extrapolated": false } ] } ], "sensitivityRange": { "min": 6.748920873941268, "max": 10.09650904372383, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "qwen3-8-max", "name": "Qwen3.8 Max (0902)", "shortName": "Qwen3.8 Max (0902)", "provider": "Alibaba", "release": "2026-09-02", "scores": { "frontierCode": 43.156299999999995, "briefcaseRubric": 57.95574288724974, "automation": 56.1905381087678, "analyst": 45, "omniscienceAccuracy": 31.683333333333337, "harveyLab": null, "itbench": 40.254237288135606, "mlcr": 20, "scicode": 52.0833333333333, "terminalBench40": 38.8888888888889, "terminalBench21": 88.7640449438202, "terminalBenchScience": 11.9047619047619, "gdpval": 1667.28, "gdpPdf": 22.8, "enterpriseOps": 47.62757385854969, "gpqa": 92.82828282828281, "hle": 43.0954587581094, "lcr": 80.3333333333333 }, "status": "scored", "index": 7.153079692825993, "discordance": 8.635484252288693, "privateEvaluations": 4, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 43.156299999999995, "expected": 40.22785198442066, "gapPoints": 0, "index": 0, "discordance": 11.379506132604623, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6090225563909775, "weight": 0.3709084741930013, "slope": 1.1394282115384504, "intercept": -24.00258573717881, "expected": 35.34263361378211, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 20, "discordance": 5, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8084310760975514, "weight": 0.6535608048002449, "slope": 0.3704444754098352, "intercept": 28.59413442622951, "expected": 43.000308469945324, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 20, "discordance": 15, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 57.95574288724974, "expected": 54.223546488763745, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.9007518796992481, "weight": 0.8113539487817287, "slope": 0.05115421850076689, "intercept": -31.304852364651993, "expected": 53.98355305730662, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7060599156233736, "weight": 0.4985206044500854, "slope": 0.9898989898989903, "intercept": 32.04444444444444, "expected": 54.614141414141415, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 19, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7369218095898536, "weight": 0.5430537534491845, "slope": 0.5548092271266981, "intercept": 36.46853627937398, "expected": 58.044450667634464, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 7, "comparablePairs": 20, "discordance": 35, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 45, "expected": 37.99421480732041, "gapPoints": 0, "index": 0, "discordance": 4.082269572635958, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6674025000565448, "weight": 0.44542609708172626, "slope": 1.3846153846153813, "intercept": -38.846153846153655, "expected": 33.269230769230745, "referenceModels": 14, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 14, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7097130242825607, "weight": 0.5036925768362985, "slope": 1.6369047619047619, "intercept": 4.851190476190476, "expected": 42.172619047619044, "referenceModels": 14, "referenceProviders": 8, "discordantPairs": 1, "comparablePairs": 13, "discordance": 7.6923076923076925, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 31.683333333333337, "expected": 44.382073752799585, "gapPoints": 12.698740419466247, "index": 28.612318771303972, "discordance": 19.08016130391419, "extrapolated": false, "predictors": [ { "key": "gpqa", "rho": 0.7608220194556027, "weight": 0.5788501452885014, "slope": 4.070991461100498, "intercept": -333.1982922201071, "expected": 44.704854522454184, "referenceModels": 11, "referenceProviders": 9, "discordantPairs": 2, "comparablePairs": 11, "discordance": 18.181818181818183, "extrapolated": false }, { "key": "hle", "rho": 0.7518796992481203, "weight": 0.5653230821414438, "slope": 1.16499634735022, "intercept": -6.1544825627922535, "expected": 44.051569477787254, "referenceModels": 20, "referenceProviders": 13, "discordantPairs": 4, "comparablePairs": 20, "discordance": 20, "extrapolated": false } ], "sensitivityRange": { "min": 22.614922452582324, "max": 30.8881615924431, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "itbench", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.43917473339246044, "weight": 0.19287444645033872, "slope": 1.2222222222222232, "intercept": 7.133333333333342, "expected": 35.00000000000003, "referenceModels": 19, "referenceProviders": 12, "discordantPairs": 6, "comparablePairs": 17, "discordance": 35.294117647058826, "extrapolated": false } ] } ], "sensitivityRange": { "min": 5.653730613145581, "max": 7.722040398110775, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "gemini-3-8-flash", "name": "Gemini 3.8 Flash (High)", "shortName": "Gemini 3.8 Flash (high)", "provider": "Google", "release": "2026-09-02", "scores": { "frontierCode": 36.3585, "briefcaseRubric": 42.12121212121212, "automation": 59.93009432118242, "analyst": null, "omniscienceAccuracy": 54.6, "harveyLab": 0, "itbench": 52.5423728813559, "mlcr": 21.6666666666667, "scicode": 56.5972222222222, "terminalBench40": 19.6969696969697, "terminalBench21": 87.6404494382023, "terminalBenchScience": 10, "gdpval": 1435.52, "gdpPdf": 21, "enterpriseOps": null, "gpqa": 95.2525252525253, "hle": 47.8220574606117, "lcr": 81.3333333333333 }, "status": "scored", "index": 5.8271371849186195, "discordance": 13.829191956508396, "privateEvaluations": 3, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 36.3585, "expected": 38.9296047971772, "gapPoints": 2.571104797177199, "index": 6.6044975554533005, "discordance": 18.82221335115136, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6571428571428571, "weight": 0.43183673469387757, "slope": 0.9029115211267607, "intercept": -8.230008705007883, "expected": 42.87227530320806, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 7, "comparablePairs": 21, "discordance": 33.333333333333336, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8209298669418919, "weight": 0.6739258464372323, "slope": 0.34973686956521643, "intercept": 29.514472463768122, "expected": 36.40322898550723, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 21, "discordance": 9.523809523809524, "extrapolated": false } ], "sensitivityRange": { "min": 2.0543059266476664, "max": 8.685805271083922, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 42.12121212121212, "expected": 47.26184203370437, "gapPoints": 5.140629912492251, "index": 10.876913999302557, "discordance": 19.00471504675403, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.8792207792207792, "weight": 0.7730291786135942, "slope": 0.0498177236674861, "intercept": -28.381978195331833, "expected": 43.13236048381781, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 21, "discordance": 4.761904761904762, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7594284704510371, "weight": 0.5767316017316018, "slope": 1.0517903762394263, "intercept": 30.709244678899946, "expected": 52.7968425799279, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 8, "comparablePairs": 21, "discordance": 38.095238095238095, "extrapolated": false } ], "sensitivityRange": { "min": 8.970682300198508, "max": 12.794954848660174, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.763081285660951, "weight": 0.58229304852597, "slope": 0.5548092271266981, "intercept": 36.46277955569554, "expected": 47.39084009000929, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ] }, { "key": "analyst", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 54.6, "expected": 46.7926334626082, "gapPoints": 0, "index": 0, "discordance": 3.6606474716197948, "extrapolated": false, "predictors": [ { "key": "gpqa", "rho": 0.6690027772631206, "weight": 0.4475647159857686, "slope": 4.005833333333251, "intercept": -331.26018518517776, "expected": 50.305555555555316, "referenceModels": 12, "referenceProviders": 9, "discordantPairs": 1, "comparablePairs": 12, "discordance": 8.333333333333334, "extrapolated": false }, { "key": "hle", "rho": 0.7558441558441559, "weight": 0.5713003879237646, "slope": 1.3886083788706713, "intercept": -22.365548532929445, "expected": 44.04056115171067, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.7643381449665965, "weight": 0.5842127998509778, "slope": 0.3759666259666258, "intercept": -3.933048433048431, "expected": 3.96225071225071, "referenceModels": 15, "referenceProviders": 9, "discordantPairs": 5, "comparablePairs": 13, "discordance": 38.46153846153846, "extrapolated": false } ] }, { "key": "itbench", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.4327686378555179, "weight": 0.1872886939113204, "slope": 1.109321882268741, "intercept": 7.4646627303632265, "expected": 30.760422258006788, "referenceModels": 20, "referenceProviders": 12, "discordantPairs": 4, "comparablePairs": 19, "discordance": 21.05263157894737, "extrapolated": false } ] } ], "sensitivityRange": { "min": 3.6749960756153914, "max": 6.458636448796175, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "grok-4-7", "name": "Grok 4.7 (Xhigh)", "shortName": "Grok 4.7 (xhigh)", "provider": "SpaceXAI", "release": "2026-09-21", "scores": { "frontierCode": 33.834199999999996, "briefcaseRubric": 55.35353535353536, "automation": 65.55835148969457, "analyst": null, "omniscienceAccuracy": 47.449999999999996, "harveyLab": 9.444444444444446, "itbench": 42.0809792843691, "mlcr": 15, "scicode": 57.4074074074074, "terminalBench40": 25.7575757575758, "terminalBench21": null, "terminalBenchScience": null, "gdpval": 1712.45, "gdpPdf": 20, "enterpriseOps": null, "gpqa": null, "hle": 43.1417979610751, "lcr": 76.6666666666667 }, "status": "scored", "index": 5.647349270155594, "discordance": 12.331758478653887, "privateEvaluations": 3, "extrapolated": true, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 33.834199999999996, "expected": 40.73565397181043, "gapPoints": 6.901453971810433, "index": 16.942047810466782, "discordance": 28.514651868318634, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6701298701298701, "weight": 0.44907404284027663, "slope": 0.8952639211267682, "intercept": -7.7183984272306105, "expected": 43.67638223004682, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 9, "comparablePairs": 21, "discordance": 42.857142857142854, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8248297712979104, "weight": 0.6803441516193631, "slope": 0.3321624705882347, "intercept": 30.238870588235308, "expected": 38.79457058823531, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 4, "comparablePairs": 21, "discordance": 19.047619047619047, "extrapolated": false } ], "sensitivityRange": { "min": 14.737005103306958, "max": 17.827950030429342, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 55.35353535353536, "expected": 54.78712057689293, "gapPoints": 0, "index": 0, "discordance": 8.480623567643029, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.8896103896103896, "weight": 0.7914066453027492, "slope": 0.05115421850076689, "intercept": -30.54711319019504, "expected": 57.05192828144322, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 21, "discordance": 14.285714285714286, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7360214285535737, "weight": 0.5417275432900434, "slope": 1.0559453335571574, "intercept": 30.35956938524345, "expected": 51.4784760563866, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0.7957231270774732, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.789730632093744, "weight": 0.6236744712671843, "slope": 0.5519596489058692, "intercept": 36.64411635156647, "expected": 50.86125882338433, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ] }, { "key": "analyst", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "omniscienceAccuracy", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "hle", "rho": 0.7636363636363637, "weight": 0.5831404958677686, "slope": 1.4701413197394517, "intercept": -26.79621920135942, "expected": 36.628320589068316, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ] }, { "key": "harveyLab", "status": "scored", "observed": 9.444444444444446, "expected": 2.155406505567955, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "gdpPdf", "rho": 0.8769096082117752, "weight": 0.7689704609741291, "slope": 0.39065214370092416, "intercept": -5.000347439371829, "expected": 2.8126954346466544, "referenceModels": 15, "referenceProviders": 9, "discordantPairs": 0, "comparablePairs": 15, "discordance": 0, "extrapolated": false }, { "key": "lcr", "rho": 0.48424862100244553, "weight": 0.23449672694277013, "slope": 0.568181818181813, "intercept": -44.191919191918785, "expected": 0, "referenceModels": 15, "referenceProviders": 9, "discordantPairs": 0, "comparablePairs": 15, "discordance": 0, "extrapolated": true } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 12, "attemptedRuns": 13 } }, { "key": "itbench", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.4442352429414547, "weight": 0.1973449510712533, "slope": 1.109321882268741, "intercept": 7.4646627303632265, "expected": 29.651100375738046, "referenceModels": 20, "referenceProviders": 12, "discordantPairs": 8, "comparablePairs": 20, "discordance": 40, "extrapolated": false } ] } ], "sensitivityRange": { "min": 4.912335034435652, "max": 5.942650010143114, "validRuns": 12, "attemptedRuns": 13 } }, { "id": "gpt-6-luna", "name": "GPT-6 Luna (Max)", "shortName": "GPT-6 Luna (max)", "provider": "OpenAI", "release": "2026-09-22", "scores": { "frontierCode": 41.9269666666667, "briefcaseRubric": 41.21212121212121, "automation": 53.16814034483175, "analyst": 43.75, "omniscienceAccuracy": 43.78333333333334, "harveyLab": 3.3333333333333335, "itbench": null, "mlcr": 16.1111111111111, "scicode": 54.6296296296296, "terminalBench40": 12.6262626262626, "terminalBench21": null, "terminalBenchScience": 8.57142857142857, "gdpval": 1430.2, "gdpPdf": 22.8, "enterpriseOps": null, "gpqa": null, "hle": 38.5078776645042, "lcr": 83.3333333333333 }, "status": "scored", "index": 5.05671390635745, "discordance": 15.30600193764918, "privateEvaluations": 3, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 41.9269666666667, "expected": 35.01231315044203, "gapPoints": 0, "index": 0, "discordance": 4.999153178502487, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.7807017543859649, "weight": 0.6094952293013235, "slope": 1.3064442352941235, "intercept": -34.120507843137595, "expected": 37.25005686274503, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 19, "discordance": 10.526315789473685, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8208958387692781, "weight": 0.6738699781087166, "slope": 0.3963659240506327, "intercept": 27.983720675105467, "expected": 32.98834092827001, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 41.21212121212121, "expected": 48.58209367230189, "gapPoints": 7.369972460180676, "index": 15.17014171907235, "discordance": 32.901719868278185, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.9315789473684211, "weight": 0.8678393351800554, "slope": 0.05120969592092242, "intercept": -31.39133056719043, "expected": 41.84877653891282, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 19, "discordance": 10.526315789473685, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.8296754734085217, "weight": 0.6883613911756545, "slope": 1.2179759318299739, "intercept": 29.301151798175884, "expected": 57.07100304389929, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 11, "comparablePairs": 18, "discordance": 61.111111111111114, "extrapolated": false } ], "sensitivityRange": { "min": 12.134987607468954, "max": 16.319454135509236, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.6760318672217583, "weight": 0.4570190854993371, "slope": 0.55634448456389, "intercept": 36.382797862856144, "expected": 43.40734943563252, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 43.75, "expected": 40.6857176495408, "gapPoints": 0, "index": 0, "discordance": 8.017132766166863, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.7933914402968284, "weight": 0.6294699775362759, "slope": 1.724210526315805, "intercept": -59.09649122807129, "expected": 35.09649122806985, "referenceModels": 13, "referenceProviders": 8, "discordantPairs": 1, "comparablePairs": 13, "discordance": 7.6923076923076925, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.8041386959705858, "weight": 0.6466390423572743, "slope": 2.237654320987654, "intercept": -4.891975308641971, "expected": 46.126543209876544, "referenceModels": 13, "referenceProviders": 8, "discordantPairs": 1, "comparablePairs": 12, "discordance": 8.333333333333334, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "hle", "rho": 0.7508771929824561, "weight": 0.5638165589412126, "slope": 1.511117505995193, "intercept": -29.098441247001865, "expected": 29.091486810551714, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 19, "discordance": 0, "extrapolated": false } ] }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6620524015840564, "weight": 0.4383133824432167, "slope": 0.33645873198112003, "intercept": -3.490560020410767, "expected": 4.18069906875877, "referenceModels": 13, "referenceProviders": 9, "discordantPairs": 2, "comparablePairs": 13, "discordance": 15.384615384615385, "extrapolated": false } ] }, { "key": "itbench", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.5408479834539814, "weight": 0.29251654120623816, "slope": 1.6908212560386415, "intercept": 0.8937198067633503, "expected": 39.44444444444438, "referenceModels": 18, "referenceProviders": 12, "discordantPairs": 11, "comparablePairs": 17, "discordance": 64.70588235294117, "extrapolated": false } ] } ], "sensitivityRange": { "min": 4.044995869156318, "max": 5.439818045169745, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "step-5", "name": "Step 5 Preview", "shortName": "Step 5 Preview", "provider": "StepFun", "release": "2026-09-18", "scores": { "frontierCode": 38.4270666666667, "briefcaseRubric": 47.77777777777778, "automation": 50.956594190962875, "analyst": 35, "omniscienceAccuracy": 41.53333333333333, "harveyLab": null, "itbench": 55.5555555555556, "mlcr": 16.6666666666667, "scicode": 58.912037037037, "terminalBench40": 33.3333333333333, "terminalBench21": null, "terminalBenchScience": 2.38095238095238, "gdpval": 1584.24, "gdpPdf": 14.799999999999999, "enterpriseOps": 47.17994628469114, "gpqa": null, "hle": 46.4782205746061, "lcr": 88.3333333333333 }, "status": "scored", "index": 4.873484103958227, "discordance": 20.37083913350459, "privateEvaluations": 3, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 38.4270666666667, "expected": 42.61442172101658, "gapPoints": 4.1873550543498865, "index": 9.826145434433448, "discordance": 29.980114256922395, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6571428571428571, "weight": 0.43183673469387757, "slope": 0.9029115211267607, "intercept": -8.230008705007883, "expected": 44.96234826877926, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 8, "comparablePairs": 21, "discordance": 38.095238095238095, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8388565773443989, "weight": 0.7036803573539595, "slope": 0.3520550352564087, "intercept": 29.438371064814838, "expected": 41.173538906695114, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 5, "comparablePairs": 20, "discordance": 25, "extrapolated": false } ], "sensitivityRange": { "min": 8.018161710404454, "max": 10.922498508362633, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 47.77777777777778, "expected": 48.765198910707745, "gapPoints": 0.9874211329299669, "index": 2.024847955071401, "discordance": 10.562392090217912, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.9, "weight": 0.81, "slope": 0.050152787861517556, "intercept": -28.924785540304732, "expected": 50.52926710142584, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 21, "discordance": 14.285714285714286, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7210669295635275, "weight": 0.5199375169101731, "slope": 1.0510398478770737, "intercept": 30.46160348738919, "expected": 46.01699323596988, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 21, "discordance": 4.761904761904762, "extrapolated": false } ], "sensitivityRange": { "min": 1.1244423170258444, "max": 4.0324559675538465, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7673815785001821, "weight": 0.5888744870214311, "slope": 0.5080458266141755, "intercept": 39.43863231558334, "expected": 56.37349320272251, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 6, "comparablePairs": 20, "discordance": 30, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 35, "expected": 35.99691991023225, "gapPoints": 0.9969199102322506, "index": 2.769458922369835, "discordance": 20.570011053373467, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.7280870326317682, "weight": 0.5301107270865335, "slope": 1.6325581395348763, "intercept": -52.96511627906935, "expected": 43.212209302325505, "referenceModels": 15, "referenceProviders": 8, "discordantPairs": 6, "comparablePairs": 15, "discordance": 40, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7076236029776369, "weight": 0.5007311634910523, "slope": 1.9239643150123051, "intercept": -0.11638638228055953, "expected": 28.358285479901554, "referenceModels": 15, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 15, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 11.517709599346151, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "hle", "rho": 0.7818181818181819, "weight": 0.6112396694214877, "slope": 1.2798909878528049, "intercept": -12.437504465880686, "expected": 47.04955117899247, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 4, "comparablePairs": 21, "discordance": 19.047619047619047, "extrapolated": false } ] }, { "key": "harveyLab", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "itbench", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.42012069200994234, "weight": 0.17650139585491284, "slope": 1.1193590196771566, "intercept": 7.35124307764813, "expected": 23.917756568870047, "referenceModels": 20, "referenceProviders": 12, "discordantPairs": 4, "comparablePairs": 20, "discordance": 20, "extrapolated": false } ] } ], "sensitivityRange": { "min": 3.0475346758100996, "max": 8.021654409494209, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "mimo-v2-6-pro", "name": "MiMo-V2.6-Pro", "shortName": "MiMo-V2.6-Pro", "provider": "Xiaomi", "release": "2026-09-21", "scores": { "frontierCode": 45.4153333333333, "briefcaseRubric": 54.74747474747475, "automation": 58.623409434190854, "analyst": 37.5, "omniscienceAccuracy": 34.849999999999994, "harveyLab": null, "itbench": null, "mlcr": 18.3333333333333, "scicode": 60.8796296296296, "terminalBench40": 34.8484848484849, "terminalBench21": null, "terminalBenchScience": 5.71428571428571, "gdpval": 1682.08, "gdpPdf": 19.2, "enterpriseOps": null, "gpqa": null, "hle": 49.351251158480096, "lcr": 86.3333333333333 }, "status": "scored", "index": 3.549197375108932, "discordance": 11.28738886755417, "privateEvaluations": 3, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 45.4153333333333, "expected": 43.499517531176515, "gapPoints": 0, "index": 0, "discordance": 3.3455627283513003, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.5974025974025974, "weight": 0.35688986338336987, "slope": 0.8896893942857217, "intercept": -7.3454729047624525, "expected": 46.81848790476178, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 21, "discordance": 9.523809523809524, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8118300901111821, "weight": 0.65906809520993, "slope": 0.34973686956521643, "intercept": 29.514472463768122, "expected": 41.70227246376811, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 54.74747474747475, "expected": 53.5014470543852, "gapPoints": 0, "index": 0, "discordance": 8.480623567643029, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.8896103896103896, "weight": 0.7914066453027492, "slope": 0.05060752655147949, "intercept": -29.661466765429957, "expected": 55.46444149628265, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 21, "discordance": 14.285714285714286, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7360214285535737, "weight": 0.5417275432900434, "slope": 1.0559453335571574, "intercept": 30.35956938524345, "expected": 50.63371978954087, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7500816044742227, "weight": 0.5626224133706242, "slope": 0.5566495166096653, "intercept": 36.36461918538064, "expected": 55.76301143086901, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 4, "comparablePairs": 21, "discordance": 19.047619047619047, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 37.5, "expected": 41.96865075264448, "gapPoints": 4.4686507526444785, "index": 10.647592125326796, "discordance": 22.03598030666818, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.7310101221050122, "weight": 0.5343757986199847, "slope": 1.6484210526315752, "intercept": -53.710526315789224, "expected": 46.64473684210523, "referenceModels": 15, "referenceProviders": 8, "discordantPairs": 6, "comparablePairs": 14, "discordance": 42.857142857142854, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7105746177708218, "weight": 0.5049162874201495, "slope": 1.869517543859649, "intercept": 1.125, "expected": 37.01973684210526, "referenceModels": 15, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 14, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 16.762660368019485, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "hle", "rho": 0.8077922077922078, "weight": 0.6525282509698095, "slope": 1.5047222601066015, "intercept": -26.102715742097928, "expected": 48.15721044017877, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 9, "comparablePairs": 21, "discordance": 42.857142857142854, "extrapolated": false } ] }, { "key": "harveyLab", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "itbench", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.4344387066255278, "weight": 0.1887369898144614, "slope": 1.1417243802067674, "intercept": 7.098514503663528, "expected": 29.019622603633458, "referenceModels": 20, "referenceProviders": 12, "discordantPairs": 5, "comparablePairs": 20, "discordance": 25, "extrapolated": false } ] } ], "sensitivityRange": { "min": 0, "max": 5.5875534560064954, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "deepseek-v4-1-flash", "name": "DeepSeek V4.1 Flash (Max)", "shortName": "DeepSeek V4.1 Flash (max)", "provider": "DeepSeek", "release": "2026-09-10", "scores": { "frontierCode": 44.0335333333333, "briefcaseRubric": 45.194274028629856, "automation": 68.89097769674058, "analyst": null, "omniscienceAccuracy": 46.400000000000006, "harveyLab": 1.6666666666666667, "itbench": 46.9161958568738, "mlcr": 22.7777777777778, "scicode": 51.851851851851805, "terminalBench40": 26.7676767676768, "terminalBench21": null, "terminalBenchScience": 9.047619047619051, "gdpval": 1600, "gdpPdf": 12.8, "enterpriseOps": null, "gpqa": null, "hle": 39.2493049119555, "lcr": 84 }, "status": "scored", "index": 3.3337154910413362, "discordance": 8.255910809584215, "privateEvaluations": 2, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 44.0335333333333, "expected": 38.042495816736114, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6558441558441559, "weight": 0.4301315567549334, "slope": 0.9944002226086934, "intercept": -14.824676541062729, "expected": 36.7368164830917, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8417293568406572, "weight": 0.7085083101673865, "slope": 0.3564935000000027, "intercept": 29.292663888888775, "expected": 38.83516666666664, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 45.194274028629856, "expected": 48.422832998364996, "gapPoints": 3.22855896973514, "index": 6.6674309820826725, "discordance": 16.51182161916843, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.9077922077922078, "weight": 0.8240866925282511, "slope": 0.05105533002993838, "intercept": -30.386912878568147, "expected": 51.30161516933327, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 5, "comparablePairs": 21, "discordance": 23.80952380952381, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7227822739820933, "weight": 0.5224142155827257, "slope": 1.0551948051948048, "intercept": 30.375180375180385, "expected": 43.88167388167389, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 20, "discordance": 5, "extrapolated": false } ], "sensitivityRange": { "min": 5.251758651724911, "max": 8.63579464398301, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.8027303132804723, "weight": 0.6443759558593651, "slope": 0.5519596489058692, "intercept": 36.64411635156647, "expected": 51.41879382227915, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ] }, { "key": "analyst", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "omniscienceAccuracy", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "hle", "rho": 0.7948051948051948, "weight": 0.6317152976893238, "slope": 1.5206420520243205, "intercept": -29.14340337567468, "expected": 30.54074018616957, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ] }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.701975996661653, "weight": 0.4927702998891211, "slope": 0.38956639566395657, "intercept": -4.986449864498645, "expected": 0, "referenceModels": 15, "referenceProviders": 9, "discordantPairs": 0, "comparablePairs": 14, "discordance": 0, "extrapolated": false } ] }, { "key": "itbench", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.445574419552664, "weight": 0.19853656335969347, "slope": 1.151761517615183, "intercept": 4.2411924119240325, "expected": 18.983739837398375, "referenceModels": 20, "referenceProviders": 12, "discordantPairs": 1, "comparablePairs": 19, "discordance": 5.2631578947368425, "extrapolated": false } ] } ], "sensitivityRange": { "min": 2.6258793258624555, "max": 4.317897321991505, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "claude-sonnet-5-5", "name": "Claude Sonnet 5.5 (Max, Default Fallback)", "shortName": "Claude Sonnet 5.5 (max with fallback)", "provider": "Anthropic", "release": "2026-09-28", "scores": { "frontierCode": 45.8291333333333, "briefcaseRubric": 65.15151515151516, "automation": 71.76882128193995, "analyst": 57.49999999999999, "omniscienceAccuracy": 53.949999999999996, "harveyLab": 2.7777777777777777, "itbench": null, "mlcr": 75, "scicode": 60.9953703703704, "terminalBench40": 63.636363636363605, "terminalBench21": null, "terminalBenchScience": 53.3333333333333, "gdpval": 1837.82, "gdpPdf": 25.8, "enterpriseOps": null, "gpqa": null, "hle": 54.9582947173309, "lcr": 82.6666666666667 }, "status": "scored", "index": 2.8479410090047046, "discordance": 3.703703703703703, "privateEvaluations": 3, "extrapolated": true, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 45.8291333333333, "expected": 50.1104844420407, "gapPoints": 4.281351108707398, "index": 8.543823027014113, "discordance": 11.111111111111109, "extrapolated": true, "predictors": [ { "key": "scicode", "rho": 0.5067079463364293, "weight": 0.2567529428804818, "slope": 0.9634935652173972, "intercept": -13.475405314010008, "expected": 45.293241545893764, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 18, "discordance": 11.11111111111111, "extrapolated": true }, { "key": "terminalBench40", "rho": 0.7758268602656214, "weight": 0.6019073171096121, "slope": 0.37290439999999875, "intercept": 28.435074444444474, "expected": 52.16535444444438, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 18, "discordance": 11.11111111111111, "extrapolated": true } ], "sensitivityRange": { "min": 6.746471755268028, "max": 11.470570361413618, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 65.15151515151516, "expected": 59.955417102713724, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "gdpval", "rho": 0.8328173374613003, "weight": 0.6935847175761293, "slope": 0.05152279828337948, "intercept": -31.8793945297885, "expected": 62.81023461137198, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": true }, { "key": "gdpPdf", "rho": 0.6697682468224858, "weight": 0.4485895044516663, "slope": 0.8632679033253143, "intercept": 33.26914148215549, "expected": 55.5414533879486, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7200417065314756, "weight": 0.5184600591447596, "slope": 0.6089098929269875, "intercept": 37.04083829479769, "expected": 75.78964966287869, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": true } ] }, { "key": "analyst", "status": "scored", "observed": 57.49999999999999, "expected": 43.80279239752813, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "scicode", "rho": 0.4588448367616168, "weight": 0.21053858422279478, "slope": 1.2652941176470573, "intercept": -36.672181372548906, "expected": 40.50490196078438, "referenceModels": 12, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 12, "discordance": 0, "extrapolated": true }, { "key": "gdpPdf", "rho": 0.8998256427215409, "weight": 0.8096861872992341, "slope": 1.596467391304348, "intercept": 3.471467391304344, "expected": 44.66032608695652, "referenceModels": 12, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 12, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 12, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "hle", "rho": 0.6718266253869969, "weight": 0.45135101457888027, "slope": 1.2138749999999952, "intercept": -12.113541666666432, "expected": 54.59895833333334, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 18, "discordance": 16.666666666666668, "extrapolated": true } ] }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6890502381874483, "weight": 0.47479023074617915, "slope": 0.3815628815628815, "intercept": -3.867521367521367, "expected": 5.976800976800976, "referenceModels": 12, "referenceProviders": 9, "discordantPairs": 3, "comparablePairs": 12, "discordance": 25, "extrapolated": false } ] }, { "key": "itbench", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] } ], "sensitivityRange": { "min": 2.3980858645052274, "max": 3.823523453804539, "validRuns": 12, "attemptedRuns": 13 } }, { "id": "kimi-k3", "name": "Kimi K3 (Max)", "shortName": "Kimi K3 (max)", "provider": "Kimi", "release": "2026-07-16", "scores": { "frontierCode": 43.6099333333333, "briefcaseRubric": 50.97236438075742, "automation": 58.272977800152034, "analyst": 38.75, "omniscienceAccuracy": 47.583333333333336, "harveyLab": 5.277777777777778, "itbench": 47.6930320150659, "mlcr": 38.3333333333333, "scicode": 59.4907407407407, "terminalBench40": 12.6262626262626, "terminalBench21": 85.0187265917603, "terminalBenchScience": null, "gdpval": 1535.63, "gdpPdf": 22, "enterpriseOps": 45.329752312742464, "gpqa": 93.53535353535351, "hle": 46.8952734012975, "lcr": 88.6666666666667 }, "status": "scored", "index": 2.7047520984125564, "discordance": 8.036765219443161, "privateEvaluations": 4, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 43.6099333333333, "expected": 37.450892203491726, "gapPoints": 0, "index": 0, "discordance": 4.753919366103723, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.6116883116883117, "weight": 0.3741625906560972, "slope": 0.8896893942857217, "intercept": -7.3454729047624525, "expected": 45.582808190476044, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 21, "discordance": 14.285714285714286, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8661470314485543, "weight": 0.750210680087143, "slope": 0.371674437704917, "intercept": 28.70229408925322, "expected": 33.395153151183976, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 50.97236438075742, "expected": 49.90133848191394, "gapPoints": 0, "index": 0, "discordance": 7.640181249707307, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.8922077922077922, "weight": 0.796034744476303, "slope": 0.05105533002993838, "intercept": -31.131674789284, "expected": 47.27042166459028, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7301696680792078, "weight": 0.5331477441829005, "slope": 1.0559453335571574, "intercept": 30.598722814248298, "expected": 53.82952015250576, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 4, "comparablePairs": 21, "discordance": 19.047619047619047, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0.5081020457179519, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.742690215263089, "weight": 0.5515887558475334, "slope": 0.5564970005867776, "intercept": 36.36485027026381, "expected": 43.39132755039987, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 20, "discordance": 0, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 38.75, "expected": 43.45096337462227, "gapPoints": 4.700963374622269, "index": 10.819008393650225, "discordance": 19.752960261961615, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.7280870326317682, "weight": 0.5301107270865335, "slope": 1.6273972602739672, "intercept": -52.61986301369835, "expected": 44.19520547945194, "referenceModels": 15, "referenceProviders": 8, "discordantPairs": 5, "comparablePairs": 15, "discordance": 33.333333333333336, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7417043341730111, "weight": 0.5501253193310297, "slope": 1.8944831829368336, "intercept": 1.0551681706316636, "expected": 42.733798195242, "referenceModels": 15, "referenceProviders": 8, "discordantPairs": 1, "comparablePairs": 15, "discordance": 6.666666666666667, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 17.884688518679308, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 47.583333333333336, "expected": 43.402865328059846, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": false, "predictors": [ { "key": "gpqa", "rho": 0.6573426573426573, "weight": 0.4320993691623062, "slope": 3.9755651815181094, "intercept": -328.52382196552605, "expected": 43.33207279061213, "referenceModels": 12, "referenceProviders": 9, "discordantPairs": 0, "comparablePairs": 11, "discordance": 0, "extrapolated": false }, { "key": "hle", "rho": 0.7493506493506493, "weight": 0.5615263956822398, "slope": 1.30133095238094, "intercept": -17.56893000573661, "expected": 43.45734079173843, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0.4301466204736582, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.7169814214682343, "weight": 0.5140623587306098, "slope": 0.36698376250615045, "intercept": -4.4320521932462205, "expected": 3.6415905818890897, "referenceModels": 15, "referenceProviders": 9, "discordantPairs": 1, "comparablePairs": 15, "discordance": 6.666666666666667, "extrapolated": false } ] }, { "key": "itbench", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.4412209240750157, "weight": 0.19467590384161074, "slope": 0.9734623015873047, "intercept": 6.594742063492027, "expected": 28.01091269841273, "referenceModels": 20, "referenceProviders": 12, "discordantPairs": 3, "comparablePairs": 20, "discordance": 15, "extrapolated": false } ] } ], "sensitivityRange": { "min": 0, "max": 4.471172129669827, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "gpt-6-astra", "name": "GPT-6 Astra (Max)", "shortName": "GPT-6 Astra (max)", "provider": "OpenAI", "release": "2026-09-03", "scores": { "frontierCode": 46.9827666666667, "briefcaseRubric": 52.02020202020202, "automation": 68.49174806688876, "analyst": 51.24999999999999, "omniscienceAccuracy": 62.6, "harveyLab": 8.611111111111109, "itbench": 48.5875706214689, "mlcr": 35, "scicode": 56.4814814814815, "terminalBench40": 59.09090909090911, "terminalBench21": 88.38951310861421, "terminalBenchScience": 63.3333333333333, "gdpval": 1571.66, "gdpPdf": 31, "enterpriseOps": null, "gpqa": 96.0606060606061, "hle": 54.6802594995366, "lcr": 80.6666666666667 }, "status": "scored", "index": 2.3495973795271943, "discordance": 11.115485074717599, "privateEvaluations": 4, "extrapolated": true, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 46.9827666666667, "expected": 45.831713518916345, "gapPoints": 0, "index": 0, "discordance": 2.763581305485599, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.7807017543859649, "weight": 0.6094952293013235, "slope": 1.3064442352941235, "intercept": -34.120507843137595, "expected": 39.6693980392157, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 19, "discordance": 0, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8208958387692781, "weight": 0.6738699781087166, "slope": 0.3963659240506327, "intercept": 27.983720675105467, "expected": 51.405343459915585, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 19, "discordance": 5.2631578947368425, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 52.02020202020202, "expected": 57.03966444015631, "gapPoints": 5.019462419954287, "index": 8.799950822327334, "discordance": 18.624613799698547, "extrapolated": true, "predictors": [ { "key": "gdpval", "rho": 0.9315789473684211, "weight": 0.8678393351800554, "slope": 0.05120969592092242, "intercept": -31.39133056719043, "expected": 49.0929001238865, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 19, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.8296754734085217, "weight": 0.6883613911756545, "slope": 1.2179759318299739, "intercept": 29.301151798175884, "expected": 67.05840568490507, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 8, "comparablePairs": 19, "discordance": 42.10526315789474, "extrapolated": true } ], "sensitivityRange": { "min": 5.417539996763545, "max": 10.61450684333748, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.6760318672217583, "weight": 0.4570190854993371, "slope": 0.55634448456389, "intercept": 36.382797862856144, "expected": 69.25769922344965, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 19, "discordance": 5.2631578947368425, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 51.24999999999999, "expected": 51.55854629199367, "gapPoints": 0.3085462919936788, "index": 0.5984386957814436, "discordance": 19.386010351237893, "extrapolated": true, "predictors": [ { "key": "scicode", "rho": 0.7933914402968284, "weight": 0.6294699775362759, "slope": 1.724210526315805, "intercept": -59.09649122807129, "expected": 38.28947368421032, "referenceModels": 13, "referenceProviders": 8, "discordantPairs": 1, "comparablePairs": 13, "discordance": 7.6923076923076925, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.8041386959705858, "weight": 0.6466390423572743, "slope": 2.237654320987654, "intercept": -4.891975308641971, "expected": 64.47530864197529, "referenceModels": 13, "referenceProviders": 8, "discordantPairs": 4, "comparablePairs": 13, "discordance": 30.76923076923077, "extrapolated": true } ], "sensitivityRange": { "min": 0, "max": 9.348827827896807, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 62.6, "expected": 53.53549236532255, "gapPoints": 0, "index": 0, "discordance": 3.687734842448359, "extrapolated": true, "predictors": [ { "key": "gpqa", "rho": 0.6690027772631206, "weight": 0.4475647159857686, "slope": 4.005833333333251, "intercept": -331.26018518517776, "expected": 53.54259259259226, "referenceModels": 12, "referenceProviders": 9, "discordantPairs": 1, "comparablePairs": 12, "discordance": 8.333333333333334, "extrapolated": true }, { "key": "hle", "rho": 0.7508771929824561, "weight": 0.5638165589412126, "slope": 1.511117505995193, "intercept": -29.098441247001865, "expected": 53.529856115107826, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 19, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6620524015840564, "weight": 0.4383133824432167, "slope": 0.33645873198112003, "intercept": -3.490560020410767, "expected": 6.939660671003955, "referenceModels": 13, "referenceProviders": 9, "discordantPairs": 2, "comparablePairs": 13, "discordance": 15.384615384615385, "extrapolated": true } ] }, { "key": "itbench", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.5408479834539814, "weight": 0.29251654120623816, "slope": 1.6908212560386415, "intercept": 0.8937198067633503, "expected": 53.30917874396124, "referenceModels": 18, "referenceProviders": 12, "discordantPairs": 8, "comparablePairs": 18, "discordance": 44.44444444444444, "extrapolated": true } ] } ], "sensitivityRange": { "min": 1.3543849991908863, "max": 4.406941319532275, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "gpt-6-1-sol", "name": "GPT-6.1 Sol (Max)", "shortName": "GPT-6.1 Sol (max)", "provider": "OpenAI", "release": "2026-09-29", "scores": { "frontierCode": 49.1209333333333, "briefcaseRubric": 55.65656565656566, "automation": 64.86844882398171, "analyst": 50, "omniscienceAccuracy": 62.1, "harveyLab": 6.944444444444444, "itbench": null, "mlcr": 33.8888888888889, "scicode": 54.1666666666667, "terminalBench40": 56.0606060606061, "terminalBench21": null, "terminalBenchScience": 58.0952380952381, "gdpval": 1593.06, "gdpPdf": 31, "enterpriseOps": null, "gpqa": null, "hle": 52.919369786839695, "lcr": 83 }, "status": "scored", "index": 1.153055546572436, "discordance": 9.077329056683952, "privateEvaluations": 3, "extrapolated": true, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 49.1209333333333, "expected": 43.764794205708334, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.7807017543859649, "weight": 0.6094952293013235, "slope": 1.3064442352941235, "intercept": -34.120507843137595, "expected": 36.64522156862747, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 19, "discordance": 0, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.8208958387692781, "weight": 0.6738699781087166, "slope": 0.3963659240506327, "intercept": 27.983720675105467, "expected": 50.2042345991561, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 19, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 55.65656565656566, "expected": 57.65080300152351, "gapPoints": 1.9942373449578454, "index": 3.4591666397173078, "discordance": 11.640383624811589, "extrapolated": true, "predictors": [ { "key": "gdpval", "rho": 0.9315789473684211, "weight": 0.8678393351800554, "slope": 0.05120969592092242, "intercept": -31.39133056719043, "expected": 50.18878761659423, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 19, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.8296754734085217, "weight": 0.6883613911756545, "slope": 1.2179759318299739, "intercept": 29.301151798175884, "expected": 67.05840568490507, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 5, "comparablePairs": 19, "discordance": 26.31578947368421, "extrapolated": true } ], "sensitivityRange": { "min": 0.011730433195957199, "max": 5.32882223505296, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.6760318672217583, "weight": 0.4570190854993371, "slope": 0.55634448456389, "intercept": 36.382797862856144, "expected": 67.57180684598333, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 19, "discordance": 10.526315789473685, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 50, "expected": 49.589781708004544, "gapPoints": 0, "index": 0, "discordance": 15.59160354524027, "extrapolated": true, "predictors": [ { "key": "scicode", "rho": 0.7933914402968284, "weight": 0.6294699775362759, "slope": 1.724210526315805, "intercept": -59.09649122807129, "expected": 34.29824561403487, "referenceModels": 13, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 13, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.8041386959705858, "weight": 0.6466390423572743, "slope": 2.237654320987654, "intercept": -4.891975308641971, "expected": 64.47530864197529, "referenceModels": 13, "referenceProviders": 8, "discordantPairs": 4, "comparablePairs": 13, "discordance": 30.76923076923077, "extrapolated": true } ], "sensitivityRange": { "min": 0, "max": 8.365392926846306, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "hle", "rho": 0.7508771929824561, "weight": 0.5638165589412126, "slope": 1.511117505995193, "intercept": -29.098441247001865, "expected": 50.86894484412469, "referenceModels": 19, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 19, "discordance": 0, "extrapolated": false } ] }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6620524015840564, "weight": 0.4383133824432167, "slope": 0.33645873198112003, "intercept": -3.490560020410767, "expected": 6.939660671003955, "referenceModels": 13, "referenceProviders": 9, "discordantPairs": 2, "comparablePairs": 13, "discordance": 15.384615384615385, "extrapolated": true } ] }, { "key": "itbench", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.5408479834539814, "weight": 0.29251654120623816, "slope": 1.6908212560386415, "intercept": 0.8937198067633503, "expected": 53.30917874396124, "referenceModels": 18, "referenceProviders": 12, "discordantPairs": 8, "comparablePairs": 18, "discordance": 44.44444444444444, "extrapolated": true } ] } ], "sensitivityRange": { "min": 0.003910144398652399, "max": 3.771109206655984, "validRuns": 13, "attemptedRuns": 13 } }, { "id": "claude-fable-5-1", "name": "Claude Fable 5.1 (Max, Default Fallback)", "shortName": "Claude Fable 5.1 (max with fallback)", "provider": "Anthropic", "release": "2026-09-01", "scores": { "frontierCode": 48.9732333333333, "briefcaseRubric": 61.51515151515151, "automation": 59.37591715646424, "analyst": 57.49999999999999, "omniscienceAccuracy": 67.23333333333333, "harveyLab": 6.388888888888888, "itbench": 49.5291902071563, "mlcr": 71.1111111111111, "scicode": 63.078703703703695, "terminalBench40": 52.020202020202, "terminalBench21": 91.3857677902622, "terminalBenchScience": 43.3333333333333, "gdpval": 1756.22, "gdpPdf": 26.200000000000003, "enterpriseOps": null, "gpqa": 93.7373737373737, "hle": 59.1288229842447, "lcr": 85.3333333333333 }, "status": "scored", "index": 0, "discordance": 0.41529965479052067, "privateEvaluations": 4, "extrapolated": true, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 48.9732333333333, "expected": 47.6742263281924, "gapPoints": 0, "index": 0, "discordance": 1.6611986191620827, "extrapolated": true, "predictors": [ { "key": "scicode", "rho": 0.5067079463364293, "weight": 0.2567529428804818, "slope": 0.9634935652173972, "intercept": -13.475405314010008, "expected": 47.300519806763305, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 18, "discordance": 5.555555555555555, "extrapolated": true }, { "key": "terminalBench40", "rho": 0.7758268602656214, "weight": 0.6019073171096121, "slope": 0.37290439999999875, "intercept": 28.435074444444474, "expected": 47.83363666666662, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 61.51515151515151, "expected": 57.53800170421539, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "gdpval", "rho": 0.8328173374613003, "weight": 0.6935847175761293, "slope": 0.05152279828337948, "intercept": -31.8793945297885, "expected": 58.60597427144822, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": true }, { "key": "gdpPdf", "rho": 0.6697682468224858, "weight": 0.4485895044516663, "slope": 0.8632679033253143, "intercept": 33.26914148215549, "expected": 55.88676054927873, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7200417065314756, "weight": 0.5184600591447596, "slope": 0.6089098929269875, "intercept": 37.04083829479769, "expected": 68.71645393695914, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 5, "comparablePairs": 18, "discordance": 27.77777777777778, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 57.49999999999999, "expected": 44.85358136906854, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "scicode", "rho": 0.4588448367616168, "weight": 0.21053858422279478, "slope": 1.2652941176470573, "intercept": -36.672181372548906, "expected": 43.14093137254903, "referenceModels": 12, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 12, "discordance": 0, "extrapolated": true }, { "key": "gdpPdf", "rho": 0.8998256427215409, "weight": 0.8096861872992341, "slope": 1.596467391304348, "intercept": 3.471467391304344, "expected": 45.298913043478265, "referenceModels": 12, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 12, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 12, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 67.23333333333333, "expected": 51.98473609649666, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "gpqa", "rho": 0.6760080419465041, "weight": 0.4569868727763465, "slope": 4.055806451612945, "intercept": -335.7779569892511, "expected": 44.402688172042986, "referenceModels": 12, "referenceProviders": 9, "discordantPairs": 0, "comparablePairs": 12, "discordance": 0, "extrapolated": false }, { "key": "hle", "rho": 0.6718266253869969, "weight": 0.45135101457888027, "slope": 1.2138749999999952, "intercept": -12.113541666666432, "expected": 59.661458333333314, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": true } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6890502381874483, "weight": 0.47479023074617915, "slope": 0.3815628815628815, "intercept": -3.867521367521367, "expected": 6.129426129426128, "referenceModels": 12, "referenceProviders": 9, "discordantPairs": 1, "comparablePairs": 12, "discordance": 8.333333333333334, "extrapolated": false } ] }, { "key": "itbench", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 12, "attemptedRuns": 13 } }, { "id": "claude-haiku-5-5", "name": "Claude Haiku 5.5 (Max)", "shortName": "Claude Haiku 5.5 (max)", "provider": "Anthropic", "release": "2026-10-07", "scores": { "frontierCode": 41.8067, "briefcaseRubric": 52.323232323232325, "automation": 35.411015132783625, "analyst": 53.75, "omniscienceAccuracy": 36.38333333333333, "harveyLab": 0.5555555555555555, "itbench": null, "mlcr": 63.3333333333333, "scicode": 54.976851851851805, "terminalBench40": 32.8282828282828, "terminalBench21": null, "terminalBenchScience": 20, "gdpval": 1620.01, "gdpPdf": 20.8, "enterpriseOps": null, "gpqa": null, "hle": 44.3929564411492, "lcr": 82.6666666666667 }, "status": "scored", "index": 0, "discordance": 8.917535853414575, "privateEvaluations": 3, "extrapolated": false, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 41.8067, "expected": 40.323315015843484, "gapPoints": 0, "index": 0, "discordance": 19.015104745192307, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.5067079463364293, "weight": 0.2567529428804818, "slope": 0.9634935652173972, "intercept": -13.475405314010008, "expected": 39.494437681159354, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 4, "comparablePairs": 18, "discordance": 22.22222222222222, "extrapolated": false }, { "key": "terminalBench40", "rho": 0.7758268602656214, "weight": 0.6019073171096121, "slope": 0.37290439999999875, "intercept": 28.435074444444474, "expected": 40.676885555555536, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 3, "comparablePairs": 17, "discordance": 17.647058823529413, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 52.323232323232325, "expected": 51.445509030318306, "gapPoints": 0, "index": 0, "discordance": 7.737502815051419, "extrapolated": false, "predictors": [ { "key": "gdpval", "rho": 0.8328173374613003, "weight": 0.6935847175761293, "slope": 0.05152279828337948, "intercept": -31.8793945297885, "expected": 51.58805391726909, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 1, "comparablePairs": 18, "discordance": 5.555555555555555, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.6697682468224858, "weight": 0.4485895044516663, "slope": 0.8632679033253143, "intercept": 33.26914148215549, "expected": 51.22511387132203, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 18, "discordance": 11.11111111111111, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7200417065314756, "weight": 0.5184600591447596, "slope": 0.6089098929269875, "intercept": 37.04083829479769, "expected": 57.030304476744234, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 7, "comparablePairs": 17, "discordance": 41.1764705882353, "extrapolated": false } ] }, { "key": "analyst", "status": "scored", "observed": 53.75, "expected": 35.89622043205647, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": false, "predictors": [ { "key": "scicode", "rho": 0.4588448367616168, "weight": 0.21053858422279478, "slope": 1.2652941176470573, "intercept": -36.672181372548906, "expected": 32.889705882352914, "referenceModels": 12, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 12, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.8998256427215409, "weight": 0.8096861872992341, "slope": 1.596467391304348, "intercept": 3.471467391304344, "expected": 36.67798913043478, "referenceModels": 12, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 12, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 12, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "hle", "rho": 0.6718266253869969, "weight": 0.45135101457888027, "slope": 1.2138749999999952, "intercept": -12.113541666666432, "expected": 41.77395833333334, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 4, "comparablePairs": 18, "discordance": 22.22222222222222, "extrapolated": false } ] }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6890502381874483, "weight": 0.47479023074617915, "slope": 0.3815628815628815, "intercept": -3.867521367521367, "expected": 4.068986568986568, "referenceModels": 12, "referenceProviders": 9, "discordantPairs": 3, "comparablePairs": 12, "discordance": 25, "extrapolated": false } ] }, { "key": "itbench", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 12, "attemptedRuns": 13 } }, { "id": "claude-opus-5-5", "name": "Claude Opus 5.5 (Max, Default Fallback)", "shortName": "Claude Opus 5.5 (max with fallback)", "provider": "Anthropic", "release": "2026-09-22", "scores": { "frontierCode": 52.17309999999999, "briefcaseRubric": 61.212121212121204, "automation": 69.5386812863607, "analyst": 56.25, "omniscienceAccuracy": 66.21666666666667, "harveyLab": 4.166666666666666, "itbench": 38.229755178907695, "mlcr": 66.6666666666667, "scicode": 66.8981481481482, "terminalBench40": 59.5959595959596, "terminalBench21": null, "terminalBenchScience": 59.047619047619, "gdpval": 1866.27, "gdpPdf": 26.200000000000003, "enterpriseOps": null, "gpqa": null, "hle": 61.353104726598694, "lcr": 84.6666666666667 }, "status": "scored", "index": 0, "discordance": 0, "privateEvaluations": 3, "extrapolated": true, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 52.17309999999999, "expected": 50.75491119531622, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "scicode", "rho": 0.5067079463364293, "weight": 0.2567529428804818, "slope": 0.9634935652173972, "intercept": -13.475405314010008, "expected": 50.98052995169091, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": true }, { "key": "terminalBench40", "rho": 0.7758268602656214, "weight": 0.6019073171096121, "slope": 0.37290439999999875, "intercept": 28.435074444444474, "expected": 50.65866999999996, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": true } ], "sensitivityRange": { "min": 0, "max": 0.1876991213865335, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 61.212121212121204, "expected": 60.981157310428294, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "gdpval", "rho": 0.8328173374613003, "weight": 0.6935847175761293, "slope": 0.05152279828337948, "intercept": -31.8793945297885, "expected": 64.27605822253412, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": true }, { "key": "gdpPdf", "rho": 0.6697682468224858, "weight": 0.4485895044516663, "slope": 0.8632679033253143, "intercept": 33.26914148215549, "expected": 55.88676054927873, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 1.5043500418428124, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7200417065314756, "weight": 0.5184600591447596, "slope": 0.6089098929269875, "intercept": 37.04083829479769, "expected": 73.32940767125453, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": true } ] }, { "key": "analyst", "status": "scored", "observed": 56.25, "expected": 45.85088527600416, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "scicode", "rho": 0.4588448367616168, "weight": 0.21053858422279478, "slope": 1.2652941176470573, "intercept": -36.672181372548906, "expected": 47.973651960784395, "referenceModels": 12, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 12, "discordance": 0, "extrapolated": true }, { "key": "gdpPdf", "rho": 0.8998256427215409, "weight": 0.8096861872992341, "slope": 1.596467391304348, "intercept": 3.471467391304344, "expected": 45.298913043478265, "referenceModels": 12, "referenceProviders": 8, "discordantPairs": 0, "comparablePairs": 12, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 12, "attemptedRuns": 13 } }, { "key": "omniscienceAccuracy", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "hle", "rho": 0.6718266253869969, "weight": 0.45135101457888027, "slope": 1.2138749999999952, "intercept": -12.113541666666432, "expected": 62.36145833333326, "referenceModels": 18, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 18, "discordance": 0, "extrapolated": true } ] }, { "key": "harveyLab", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "gdpPdf", "rho": 0.6890502381874483, "weight": 0.47479023074617915, "slope": 0.3815628815628815, "intercept": -3.867521367521367, "expected": 6.129426129426128, "referenceModels": 12, "referenceProviders": 9, "discordantPairs": 2, "comparablePairs": 12, "discordance": 16.666666666666668, "extrapolated": false } ] }, { "key": "itbench", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] }, { "key": "mlcr", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [] } ], "sensitivityRange": { "min": 0, "max": 0.5014500139476041, "validRuns": 12, "attemptedRuns": 13 } }, { "id": "k2-horizon-375b-a23b", "name": "K2 Horizon 375B A23B", "shortName": "K2 Horizon 375B A23B", "provider": "Institute of Foundation Models", "release": "2026-09-03", "scores": { "frontierCode": 32.0955, "briefcaseRubric": 41.64926931106472, "automation": 37.20802754388014, "analyst": null, "omniscienceAccuracy": 18.183333333333334, "harveyLab": null, "itbench": null, "mlcr": null, "scicode": 42.939814814814795, "terminalBench40": 1.51515151515152, "terminalBench21": 71.91011235955061, "terminalBenchScience": null, "gdpval": 1360.51, "gdpPdf": 7.3999999999999995, "enterpriseOps": null, "gpqa": 87.27272727272731, "hle": 31.974050046339197, "lcr": 80 }, "status": "scored", "index": 0, "discordance": 4.386736077492109, "privateEvaluations": 3, "extrapolated": true, "slices": [ { "key": "frontierCode", "status": "scored", "observed": 32.0955, "expected": 28.45164728917144, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "scicode", "rho": 0.5818181818181818, "weight": 0.33851239669421485, "slope": 1.0987857556561036, "intercept": -21.33363226495699, "expected": 25.848024604072435, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": true }, { "key": "terminalBench40", "rho": 0.7988304089244538, "weight": 0.63813002222241, "slope": 0.3564935000000027, "intercept": 29.292663888888775, "expected": 29.83280555555545, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "briefcaseRubric", "status": "scored", "observed": 41.64926931106472, "expected": 38.01741021333385, "gapPoints": 0, "index": 0, "discordance": 0, "extrapolated": true, "predictors": [ { "key": "gdpval", "rho": 0.8753246753246753, "weight": 0.7661932872322482, "slope": 0.05115421850076689, "intercept": -31.298013108295216, "expected": 38.29781270418314, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false }, { "key": "gdpPdf", "rho": 0.7152151690891616, "weight": 0.511532738095238, "slope": 1.0989417670452148, "intercept": 29.465243568691857, "expected": 37.59741264482645, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": true } ], "sensitivityRange": { "min": 0, "max": 0, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "automation", "status": "unscored", "reason": "Two supported public predictors required", "predictors": [ { "key": "terminalBench40", "rho": 0.7136824971513834, "weight": 0.5093427067402344, "slope": 0.4596118018594717, "intercept": 42.52079752724631, "expected": 43.21717904521521, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 0, "comparablePairs": 21, "discordance": 0, "extrapolated": false } ] }, { "key": "analyst", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "omniscienceAccuracy", "status": "scored", "observed": 18.183333333333334, "expected": 17.69204933126414, "gapPoints": 0, "index": 0, "discordance": 13.160208232476325, "extrapolated": false, "predictors": [ { "key": "gpqa", "rho": 0.7670764828304891, "weight": 0.5884063305115936, "slope": 5.9465999999998305, "intercept": -511.6919999999842, "expected": 7.284000000001242, "referenceModels": 12, "referenceProviders": 9, "discordantPairs": 2, "comparablePairs": 12, "discordance": 16.666666666666668, "extrapolated": false }, { "key": "hle", "rho": 0.7532467532467533, "weight": 0.5673806712767752, "slope": 1.2798909878528049, "intercept": -12.437504465880686, "expected": 28.48579403358341, "referenceModels": 21, "referenceProviders": 13, "discordantPairs": 2, "comparablePairs": 21, "discordance": 9.523809523809524, "extrapolated": false } ], "sensitivityRange": { "min": 0, "max": 26.15091580114879, "validRuns": 13, "attemptedRuns": 13 } }, { "key": "harveyLab", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "itbench", "status": "unscored", "reason": "Private anchor unavailable" }, { "key": "mlcr", "status": "unscored", "reason": "Private anchor unavailable" } ], "sensitivityRange": { "min": 0, "max": 8.716971933716264, "validRuns": 13, "attemptedRuns": 13 } } ] }