Files

9387 lines
182 KiB
JSON

{
"sources": {
"kai": {
"model": "kai",
"weights": "8069bf5cc5e897ce28399d3a5d2a17097e3859d85e5a4cda50c7f19f1e60e9f4",
"device": "cpu",
"dtype": "as served"
},
"laya": {
"bundle": "hanzoai/kai-1",
"revision": "b50502c28537df49a3621f6fa543f9e8521e8a9c",
"runtime": "laya 0.3.20",
"device": "cpu",
"dtype": "f32"
},
"jev": {
"model": "typesafe/jev-1.13",
"served": [
"typesafe/jev-1.13-20260917"
]
}
},
"suites": {
"jev.ag_news": {
"kai": {
"n": 397,
"accuracy": 0.9421,
"macro_f1": 0.9356,
"ece": 0.0384,
"brier": 0.0897,
"nll": 0.1749,
"mean_confidence": 0.9294,
"acc_at_50_coverage": 1.0,
"unanswered": 3,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0077,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0067,
"0.9": 0.0252,
"1.0": 0.0579
}
},
"laya": {
"n": 400,
"accuracy": 0.95,
"macro_f1": 0.9439,
"ece": 0.0315,
"brier": 0.0808,
"nll": 0.1636,
"mean_confidence": 0.9221,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0067,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0067,
"0.9": 0.025,
"1.0": 0.05
}
},
"laya:english": {
"n": 400,
"accuracy": 0.95,
"macro_f1": 0.9439,
"ece": 0.0316,
"brier": 0.0808,
"nll": 0.1637,
"mean_confidence": 0.922,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0067,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0067,
"0.9": 0.025,
"1.0": 0.05
}
},
"laya:multilingual": {
"n": 400,
"accuracy": 0.935,
"macro_f1": 0.9289,
"ece": 0.038,
"brier": 0.0978,
"nll": 0.2395,
"mean_confidence": 0.973,
"acc_at_50_coverage": 0.985,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0192,
"risk_at_coverage": {
"0.1": 0.025,
"0.25": 0.01,
"0.5": 0.015,
"0.75": 0.0167,
"0.9": 0.025,
"1.0": 0.065
}
},
"laya:typed-decisions": {
"n": 400,
"accuracy": 0.9525,
"macro_f1": 0.9468,
"ece": 0.155,
"brier": 0.1137,
"nll": 0.2801,
"mean_confidence": 0.7975,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.006,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0033,
"0.9": 0.025,
"1.0": 0.0475
}
},
"jev": {
"n": 400,
"accuracy": 0.86,
"macro_f1": 0.8485,
"ece": 0.1023,
"brier": 0.2323,
"nll": 1.805,
"mean_confidence": 0.9569,
"acc_at_50_coverage": 0.925,
"unanswered": 0,
"zero_prob": 0.0575,
"questions": 400,
"aurc": 0.0832,
"risk_at_coverage": {
"0.1": 0.0772,
"0.25": 0.0772,
"0.5": 0.0772,
"0.75": 0.0783,
"0.9": 0.1028,
"1.0": 0.14
},
"latency_p50_ms": 232.8,
"latency_p95_ms": 811.9,
"cost_usd": 0.00685,
"input_tokens": 163075,
"served": {
"typesafe/jev-1.13-20260917": 400
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.9421,
"kai~jev": 0.864,
"laya~jev": 0.8525
}
},
"jev.emotion": {
"kai": {
"n": 399,
"accuracy": 0.9373,
"macro_f1": 0.8949,
"ece": 0.025,
"brier": 0.0801,
"nll": 0.145,
"mean_confidence": 0.9471,
"acc_at_50_coverage": 1.0,
"unanswered": 1,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0047,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0195,
"1.0": 0.0627
}
},
"laya": {
"n": 400,
"accuracy": 0.595,
"macro_f1": 0.4752,
"ece": 0.3063,
"brier": 0.6963,
"nll": 2.0194,
"mean_confidence": 0.9013,
"acc_at_50_coverage": 0.715,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2655,
"risk_at_coverage": {
"0.1": 0.15,
"0.25": 0.2,
"0.5": 0.285,
"0.75": 0.3533,
"0.9": 0.3833,
"1.0": 0.405
}
},
"laya:english": {
"n": 400,
"accuracy": 0.595,
"macro_f1": 0.4752,
"ece": 0.3063,
"brier": 0.6963,
"nll": 2.0194,
"mean_confidence": 0.9013,
"acc_at_50_coverage": 0.715,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2655,
"risk_at_coverage": {
"0.1": 0.15,
"0.25": 0.2,
"0.5": 0.285,
"0.75": 0.3533,
"0.9": 0.3833,
"1.0": 0.405
}
},
"laya:multilingual": {
"n": 400,
"accuracy": 0.5375,
"macro_f1": 0.4342,
"ece": 0.3323,
"brier": 0.7802,
"nll": 2.2149,
"mean_confidence": 0.8681,
"acc_at_50_coverage": 0.64,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.3246,
"risk_at_coverage": {
"0.1": 0.15,
"0.25": 0.24,
"0.5": 0.36,
"0.75": 0.4267,
"0.9": 0.4389,
"1.0": 0.4625
}
},
"laya:typed-decisions": {
"n": 400,
"accuracy": 0.6,
"macro_f1": 0.4741,
"ece": 0.2005,
"brier": 0.6113,
"nll": 1.3242,
"mean_confidence": 0.7928,
"acc_at_50_coverage": 0.71,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2631,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.17,
"0.5": 0.29,
"0.75": 0.3433,
"0.9": 0.3778,
"1.0": 0.4
}
},
"jev": {
"n": 400,
"accuracy": 0.6025,
"macro_f1": 0.5155,
"ece": 0.2795,
"brier": 0.6464,
"nll": 4.6832,
"mean_confidence": 0.8687,
"acc_at_50_coverage": 0.785,
"unanswered": 0,
"zero_prob": 0.1425,
"questions": 400,
"aurc": 0.2311,
"risk_at_coverage": {
"0.1": 0.1269,
"0.25": 0.1269,
"0.5": 0.2175,
"0.75": 0.3308,
"0.9": 0.375,
"1.0": 0.3975
},
"latency_p50_ms": 231.1,
"latency_p95_ms": 449.7,
"cost_usd": 0.00652,
"input_tokens": 155266,
"served": {
"typesafe/jev-1.13-20260917": 400
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.6065,
"kai~jev": 0.604,
"laya~jev": 0.68
}
},
"jev.banking77_full": {
"kai": {
"n": 400,
"accuracy": 0.9125,
"macro_f1": 0.3919,
"ece": 0.0788,
"brier": 0.1672,
"nll": 0.8039,
"mean_confidence": 0.976,
"acc_at_50_coverage": 0.975,
"unanswered": 0,
"zero_prob": 0.015,
"questions": 400,
"aurc": 0.0256,
"risk_at_coverage": {
"0.1": 0.025,
"0.25": 0.01,
"0.5": 0.025,
"0.75": 0.0267,
"0.9": 0.0556,
"1.0": 0.0875
}
},
"laya": {
"n": 400,
"accuracy": 0.425,
"macro_f1": 0.1121,
"ece": 0.3809,
"brier": 0.9246,
"nll": 4.7967,
"mean_confidence": 0.8005,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.0975,
"questions": 400,
"aurc": 0.3559,
"risk_at_coverage": {
"0.1": 0.075,
"0.25": 0.24,
"0.5": 0.4,
"0.75": 0.4967,
"0.9": 0.5417,
"1.0": 0.575
}
},
"laya:english": {
"n": 400,
"accuracy": 0.425,
"macro_f1": 0.1121,
"ece": 0.3809,
"brier": 0.9246,
"nll": 4.7967,
"mean_confidence": 0.8005,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.0975,
"questions": 400,
"aurc": 0.3559,
"risk_at_coverage": {
"0.1": 0.075,
"0.25": 0.24,
"0.5": 0.4,
"0.75": 0.4967,
"0.9": 0.5417,
"1.0": 0.575
}
},
"laya:multilingual": {
"n": 400,
"accuracy": 0.47,
"macro_f1": 0.1138,
"ece": 0.3851,
"brier": 0.8859,
"nll": 4.251,
"mean_confidence": 0.8524,
"acc_at_50_coverage": 0.65,
"unanswered": 0,
"zero_prob": 0.07,
"questions": 400,
"aurc": 0.3111,
"risk_at_coverage": {
"0.1": 0.05,
"0.25": 0.12,
"0.5": 0.35,
"0.75": 0.46,
"0.9": 0.5056,
"1.0": 0.53
}
},
"laya:typed-decisions": {
"n": 400,
"accuracy": 0.4925,
"macro_f1": 0.1307,
"ece": 0.0946,
"brier": 0.6744,
"nll": 2.0952,
"mean_confidence": 0.4518,
"acc_at_50_coverage": 0.64,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2984,
"risk_at_coverage": {
"0.1": 0.025,
"0.25": 0.17,
"0.5": 0.36,
"0.75": 0.43,
"0.9": 0.4778,
"1.0": 0.5075
}
},
"jev": {
"n": 400,
"accuracy": 0.835,
"macro_f1": 0.3137,
"ece": 0.0729,
"brier": 0.257,
"nll": 1.3471,
"mean_confidence": 0.881,
"acc_at_50_coverage": 0.965,
"unanswered": 0,
"zero_prob": 0.035,
"questions": 400,
"aurc": 0.0579,
"risk_at_coverage": {
"0.1": 0.0253,
"0.25": 0.0253,
"0.5": 0.0321,
"0.75": 0.0967,
"0.9": 0.1157,
"1.0": 0.165
},
"latency_p50_ms": 235.7,
"latency_p95_ms": 534.8,
"cost_usd": 0.02723,
"input_tokens": 648436,
"served": {
"typesafe/jev-1.13-20260917": 400
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.44,
"kai~jev": 0.8325,
"laya~jev": 0.4675
}
},
"app.support_triage": {
"kai": {
"n": 400,
"accuracy": 0.4325,
"macro_f1": 0.4137,
"ece": 0.4609,
"brier": 0.9946,
"nll": 4.1652,
"mean_confidence": 0.8934,
"acc_at_50_coverage": 0.56,
"unanswered": 0,
"zero_prob": 0.05,
"questions": 400,
"aurc": 0.4138,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.31,
"0.5": 0.44,
"0.75": 0.4933,
"0.9": 0.5361,
"1.0": 0.5675
}
},
"laya": {
"n": 400,
"accuracy": 0.5025,
"macro_f1": 0.4802,
"ece": 0.0907,
"brier": 0.6438,
"nll": 1.4237,
"mean_confidence": 0.5065,
"acc_at_50_coverage": 0.595,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.3208,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.19,
"0.5": 0.405,
"0.75": 0.4567,
"0.9": 0.4806,
"1.0": 0.4975
}
},
"laya:english": {
"n": 400,
"accuracy": 0.5025,
"macro_f1": 0.4802,
"ece": 0.0907,
"brier": 0.6438,
"nll": 1.4237,
"mean_confidence": 0.5065,
"acc_at_50_coverage": 0.595,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.3208,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.19,
"0.5": 0.405,
"0.75": 0.4567,
"0.9": 0.4806,
"1.0": 0.4975
}
},
"laya:multilingual": {
"n": 400,
"accuracy": 0.54,
"macro_f1": 0.5418,
"ece": 0.0862,
"brier": 0.5721,
"nll": 1.2147,
"mean_confidence": 0.611,
"acc_at_50_coverage": 0.745,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2401,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.1,
"0.5": 0.255,
"0.75": 0.3933,
"0.9": 0.4333,
"1.0": 0.46
}
},
"laya:typed-decisions": {
"n": 400,
"accuracy": 0.505,
"macro_f1": 0.4699,
"ece": 0.1149,
"brier": 0.6603,
"nll": 1.4739,
"mean_confidence": 0.4165,
"acc_at_50_coverage": 0.595,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.3274,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.22,
"0.5": 0.405,
"0.75": 0.4467,
"0.9": 0.4694,
"1.0": 0.495
}
},
"jev": {
"n": 400,
"accuracy": 0.365,
"macro_f1": 0.3682,
"ece": 0.4827,
"brier": 1.0557,
"nll": 10.1202,
"mean_confidence": 0.8392,
"acc_at_50_coverage": 0.48,
"unanswered": 0,
"zero_prob": 0.335,
"questions": 400,
"aurc": 0.4868,
"risk_at_coverage": {
"0.1": 0.2632,
"0.25": 0.3714,
"0.5": 0.5217,
"0.75": 0.61,
"0.9": 0.6392,
"1.0": 0.635
},
"latency_p50_ms": 235.7,
"latency_p95_ms": 684.4,
"cost_usd": 0.01011,
"input_tokens": 240700,
"served": {
"typesafe/jev-1.13-20260917": 400
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.4975,
"kai~jev": 0.39,
"laya~jev": 0.6125
}
},
"app.email_spam": {
"kai": {
"n": 399,
"accuracy": 0.995,
"macro_f1": 0.995,
"ece": 0.0051,
"brier": 0.0095,
"nll": 0.0188,
"mean_confidence": 0.9991,
"acc_at_50_coverage": 1.0,
"unanswered": 1,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0001,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.005
}
},
"laya": {
"n": 400,
"accuracy": 0.9975,
"macro_f1": 0.9975,
"ece": 0.0108,
"brier": 0.0052,
"nll": 0.0126,
"mean_confidence": 0.9892,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.0025
}
},
"laya:english": {
"n": 400,
"accuracy": 0.9975,
"macro_f1": 0.9975,
"ece": 0.0109,
"brier": 0.0052,
"nll": 0.0127,
"mean_confidence": 0.9892,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.0025
}
},
"laya:multilingual": {
"n": 400,
"accuracy": 0.9975,
"macro_f1": 0.9975,
"ece": 0.0026,
"brier": 0.0049,
"nll": 0.0116,
"mean_confidence": 0.9996,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.0025
}
},
"laya:typed-decisions": {
"n": 400,
"accuracy": 0.9675,
"macro_f1": 0.9675,
"ece": 0.1144,
"brier": 0.0888,
"nll": 0.1819,
"mean_confidence": 0.8531,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0017,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0056,
"1.0": 0.0325
}
},
"jev": {
"n": 400,
"accuracy": 0.9775,
"macro_f1": 0.9774,
"ece": 0.0604,
"brier": 0.0379,
"nll": 0.0967,
"mean_confidence": 0.9247,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0007,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.0225
},
"latency_p50_ms": 230.2,
"latency_p95_ms": 343.4,
"cost_usd": 0.00999,
"input_tokens": 237809,
"served": {
"typesafe/jev-1.13-20260917": 400
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.9925,
"kai~jev": 0.9774,
"laya~jev": 0.98
}
},
"app.phishing": {
"kai": {
"n": 400,
"accuracy": 0.9875,
"macro_f1": 0.9868,
"ece": 0.0125,
"brier": 0.025,
"nll": 0.0908,
"mean_confidence": 1.0,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0003,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.0125
}
},
"laya": {
"n": 400,
"accuracy": 0.9825,
"macro_f1": 0.9815,
"ece": 0.0101,
"brier": 0.026,
"nll": 0.0439,
"mean_confidence": 0.9743,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0005,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.0175
}
},
"laya:english": {
"n": 400,
"accuracy": 0.98,
"macro_f1": 0.9789,
"ece": 0.0121,
"brier": 0.0292,
"nll": 0.048,
"mean_confidence": 0.9738,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0007,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.02
}
},
"laya:multilingual": {
"n": 400,
"accuracy": 0.9925,
"macro_f1": 0.9921,
"ece": 0.0063,
"brier": 0.0138,
"nll": 0.0243,
"mean_confidence": 0.9981,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0002,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.0075
}
},
"laya:typed-decisions": {
"n": 400,
"accuracy": 0.94,
"macro_f1": 0.9353,
"ece": 0.0762,
"brier": 0.1062,
"nll": 0.2039,
"mean_confidence": 0.865,
"acc_at_50_coverage": 0.99,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0146,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.02,
"0.5": 0.01,
"0.75": 0.0067,
"0.9": 0.0265,
"1.0": 0.06
}
},
"jev": {
"n": 400,
"accuracy": 0.9,
"macro_f1": 0.8874,
"ece": 0.0388,
"brier": 0.1523,
"nll": 0.2495,
"mean_confidence": 0.887,
"acc_at_50_coverage": 0.995,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0208,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.005,
"0.75": 0.05,
"0.9": 0.075,
"1.0": 0.1
},
"latency_p50_ms": 232.8,
"latency_p95_ms": 390.9,
"cost_usd": 0.01065,
"input_tokens": 253672,
"served": {
"typesafe/jev-1.13-20260917": 400
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.995,
"kai~jev": 0.8925,
"laya~jev": 0.8925
}
},
"app.guardrails_jailbreak": {
"kai": {
"n": 399,
"accuracy": 0.9198,
"macro_f1": 0.8749,
"ece": 0.077,
"brier": 0.1492,
"nll": 0.4208,
"mean_confidence": 0.9897,
"acc_at_50_coverage": 0.9749,
"unanswered": 1,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.0249,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0101,
"0.5": 0.0251,
"0.75": 0.0268,
"0.9": 0.0557,
"1.0": 0.0802
}
},
"laya": {
"n": 400,
"accuracy": 0.705,
"macro_f1": 0.6818,
"ece": 0.2617,
"brier": 0.5427,
"nll": 1.9121,
"mean_confidence": 0.9658,
"acc_at_50_coverage": 0.82,
"unanswered": 0,
"zero_prob": 0.0025,
"questions": 400,
"aurc": 0.2304,
"risk_at_coverage": {
"0.1": 0.225,
"0.25": 0.18,
"0.5": 0.18,
"0.75": 0.2533,
"0.9": 0.275,
"1.0": 0.295
}
},
"laya:english": {
"n": 400,
"accuracy": 0.7075,
"macro_f1": 0.684,
"ece": 0.2591,
"brier": 0.5389,
"nll": 1.8576,
"mean_confidence": 0.9657,
"acc_at_50_coverage": 0.83,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.225,
"risk_at_coverage": {
"0.1": 0.225,
"0.25": 0.18,
"0.5": 0.17,
"0.75": 0.2467,
"0.9": 0.275,
"1.0": 0.2925
}
},
"laya:multilingual": {
"n": 400,
"accuracy": 0.805,
"macro_f1": 0.7667,
"ece": 0.1528,
"brier": 0.3347,
"nll": 1.0979,
"mean_confidence": 0.9578,
"acc_at_50_coverage": 0.845,
"unanswered": 0,
"zero_prob": 0.0025,
"questions": 400,
"aurc": 0.1372,
"risk_at_coverage": {
"0.1": 0.025,
"0.25": 0.17,
"0.5": 0.155,
"0.75": 0.1367,
"0.9": 0.1639,
"1.0": 0.195
}
},
"laya:typed-decisions": {
"n": 400,
"accuracy": 0.7625,
"macro_f1": 0.733,
"ece": 0.0869,
"brier": 0.3365,
"nll": 0.5061,
"mean_confidence": 0.841,
"acc_at_50_coverage": 0.885,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.1134,
"risk_at_coverage": {
"0.1": 0.025,
"0.25": 0.04,
"0.5": 0.115,
"0.75": 0.18,
"0.9": 0.2083,
"1.0": 0.2375
}
},
"jev": {
"n": 400,
"accuracy": 0.94,
"macro_f1": 0.9183,
"ece": 0.0469,
"brier": 0.1044,
"nll": 0.1891,
"mean_confidence": 0.8963,
"acc_at_50_coverage": 0.995,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.012,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0037,
"0.75": 0.021,
"0.9": 0.0375,
"1.0": 0.06
},
"latency_p50_ms": 235.1,
"latency_p95_ms": 375.8,
"cost_usd": 0.00615,
"input_tokens": 146325,
"served": {
"typesafe/jev-1.13-20260917": 400
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.6642,
"kai~jev": 0.9148,
"laya~jev": 0.74
}
},
"app.moderation_toxicity": {
"kai": {
"n": 400,
"accuracy": 0.8,
"macro_f1": 0.7951,
"ece": 0.1953,
"brier": 0.3856,
"nll": 1.8034,
"mean_confidence": 0.9916,
"acc_at_50_coverage": 0.84,
"unanswered": 0,
"zero_prob": 0.025,
"questions": 400,
"aurc": 0.1396,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.1,
"0.5": 0.16,
"0.75": 0.1533,
"0.9": 0.1722,
"1.0": 0.2
}
},
"laya": {
"n": 400,
"accuracy": 0.53,
"macro_f1": 0.4003,
"ece": 0.2967,
"brier": 0.6015,
"nll": 0.8856,
"mean_confidence": 0.8248,
"acc_at_50_coverage": 0.735,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2811,
"risk_at_coverage": {
"0.1": 0.125,
"0.25": 0.16,
"0.5": 0.265,
"0.75": 0.3833,
"0.9": 0.4417,
"1.0": 0.47
}
},
"laya:english": {
"n": 400,
"accuracy": 0.53,
"macro_f1": 0.4003,
"ece": 0.2961,
"brier": 0.6015,
"nll": 0.8862,
"mean_confidence": 0.8242,
"acc_at_50_coverage": 0.735,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2811,
"risk_at_coverage": {
"0.1": 0.15,
"0.25": 0.18,
"0.5": 0.265,
"0.75": 0.3833,
"0.9": 0.4417,
"1.0": 0.47
}
},
"laya:multilingual": {
"n": 400,
"accuracy": 0.535,
"macro_f1": 0.4136,
"ece": 0.3729,
"brier": 0.7377,
"nll": 1.2217,
"mean_confidence": 0.9079,
"acc_at_50_coverage": 0.71,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2516,
"risk_at_coverage": {
"0.1": 0.05,
"0.25": 0.09,
"0.5": 0.29,
"0.75": 0.39,
"0.9": 0.4472,
"1.0": 0.465
}
},
"laya:typed-decisions": {
"n": 400,
"accuracy": 0.53,
"macro_f1": 0.3967,
"ece": 0.244,
"brier": 0.5315,
"nll": 0.7305,
"mean_confidence": 0.7553,
"acc_at_50_coverage": 0.72,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2713,
"risk_at_coverage": {
"0.1": 0.125,
"0.25": 0.15,
"0.5": 0.28,
"0.75": 0.3667,
"0.9": 0.4472,
"1.0": 0.47
}
},
"jev": {
"n": 400,
"accuracy": 0.6625,
"macro_f1": 0.6285,
"ece": 0.1768,
"brier": 0.4316,
"nll": 0.6192,
"mean_confidence": 0.8393,
"acc_at_50_coverage": 0.86,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.1382,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.005,
"0.5": 0.1422,
"0.75": 0.2393,
"0.9": 0.3111,
"1.0": 0.3375
},
"latency_p50_ms": 232.3,
"latency_p95_ms": 886.7,
"cost_usd": 0.00597,
"input_tokens": 142041,
"served": {
"typesafe/jev-1.13-20260917": 400
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.685,
"kai~jev": 0.7125,
"laya~jev": 0.8125
}
},
"app.rag_relevance": {
"kai": {
"n": 400,
"accuracy": 0.6725,
"macro_f1": 0.6705,
"ece": 0.0232,
"brier": 0.4201,
"nll": 0.6091,
"mean_confidence": 0.6922,
"acc_at_50_coverage": 0.76,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2332,
"risk_at_coverage": {
"0.1": 0.175,
"0.25": 0.18,
"0.5": 0.24,
"0.75": 0.2867,
"0.9": 0.3083,
"1.0": 0.3275
}
},
"laya": {
"n": 400,
"accuracy": 0.625,
"macro_f1": 0.6234,
"ece": 0.1175,
"brier": 0.4656,
"nll": 0.671,
"mean_confidence": 0.7425,
"acc_at_50_coverage": 0.72,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2795,
"risk_at_coverage": {
"0.1": 0.175,
"0.25": 0.2,
"0.5": 0.28,
"0.75": 0.32,
"0.9": 0.35,
"1.0": 0.375
}
},
"laya:english": {
"n": 400,
"accuracy": 0.625,
"macro_f1": 0.6234,
"ece": 0.1175,
"brier": 0.4656,
"nll": 0.671,
"mean_confidence": 0.7425,
"acc_at_50_coverage": 0.72,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2795,
"risk_at_coverage": {
"0.1": 0.175,
"0.25": 0.2,
"0.5": 0.28,
"0.75": 0.32,
"0.9": 0.35,
"1.0": 0.375
}
},
"laya:multilingual": {
"n": 400,
"accuracy": 0.6725,
"macro_f1": 0.6712,
"ece": 0.1689,
"brier": 0.4859,
"nll": 0.7745,
"mean_confidence": 0.8376,
"acc_at_50_coverage": 0.76,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2619,
"risk_at_coverage": {
"0.1": 0.175,
"0.25": 0.22,
"0.5": 0.24,
"0.75": 0.2933,
"0.9": 0.3111,
"1.0": 0.3275
}
},
"laya:typed-decisions": {
"n": 400,
"accuracy": 0.625,
"macro_f1": 0.6239,
"ece": 0.0419,
"brier": 0.4458,
"nll": 0.6367,
"mean_confidence": 0.6433,
"acc_at_50_coverage": 0.73,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.2798,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.24,
"0.5": 0.27,
"0.75": 0.33,
"0.9": 0.3556,
"1.0": 0.375
}
},
"jev": {
"n": 400,
"accuracy": 0.62,
"macro_f1": 0.5924,
"ece": 0.2788,
"brier": 0.6282,
"nll": 1.1129,
"mean_confidence": 0.881,
"acc_at_50_coverage": 0.67,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 400,
"aurc": 0.3309,
"risk_at_coverage": {
"0.1": 0.2824,
"0.25": 0.2963,
"0.5": 0.3237,
"0.75": 0.3667,
"0.9": 0.3847,
"1.0": 0.38
},
"latency_p50_ms": 234.6,
"latency_p95_ms": 528.5,
"cost_usd": 0.00669,
"input_tokens": 159206,
"served": {
"typesafe/jev-1.13-20260917": 400
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.7275,
"kai~jev": 0.6825,
"laya~jev": 0.67
}
},
"app.model_routing_domain": {
"kai": {
"n": 399,
"accuracy": 1.0,
"macro_f1": 1.0,
"ece": 0.0004,
"brier": 0.0,
"nll": 0.0004,
"mean_confidence": 0.9996,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 399,
"aurc": 0.0,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.0
}
},
"laya": {
"n": 399,
"accuracy": 0.6391,
"macro_f1": 0.3199,
"ece": 0.0889,
"brier": 0.4938,
"nll": 1.089,
"mean_confidence": 0.7121,
"acc_at_50_coverage": 0.8744,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 399,
"aurc": 0.1758,
"risk_at_coverage": {
"0.1": 0.1026,
"0.25": 0.0505,
"0.5": 0.1256,
"0.75": 0.2642,
"0.9": 0.3287,
"1.0": 0.3609
}
},
"laya:english": {
"n": 399,
"accuracy": 0.6391,
"macro_f1": 0.3199,
"ece": 0.0889,
"brier": 0.4938,
"nll": 1.089,
"mean_confidence": 0.7121,
"acc_at_50_coverage": 0.8744,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 399,
"aurc": 0.1758,
"risk_at_coverage": {
"0.1": 0.1026,
"0.25": 0.0505,
"0.5": 0.1256,
"0.75": 0.2642,
"0.9": 0.3287,
"1.0": 0.3609
}
},
"laya:multilingual": {
"n": 399,
"accuracy": 0.4411,
"macro_f1": 0.2362,
"ece": 0.1957,
"brier": 0.7607,
"nll": 1.7542,
"mean_confidence": 0.6316,
"acc_at_50_coverage": 0.593,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 399,
"aurc": 0.4835,
"risk_at_coverage": {
"0.1": 0.5641,
"0.25": 0.4141,
"0.5": 0.407,
"0.75": 0.4783,
"0.9": 0.532,
"1.0": 0.5589
}
},
"laya:typed-decisions": {
"n": 399,
"accuracy": 0.6591,
"macro_f1": 0.3426,
"ece": 0.1058,
"brier": 0.503,
"nll": 1.0623,
"mean_confidence": 0.5909,
"acc_at_50_coverage": 0.8342,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 399,
"aurc": 0.2484,
"risk_at_coverage": {
"0.1": 0.2308,
"0.25": 0.2121,
"0.5": 0.1658,
"0.75": 0.2709,
"0.9": 0.3092,
"1.0": 0.3409
}
},
"jev": {
"n": 399,
"accuracy": 0.9774,
"macro_f1": 0.5915,
"ece": 0.0138,
"brier": 0.04,
"nll": 0.0847,
"mean_confidence": 0.9685,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 399,
"aurc": 0.0028,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0061,
"0.9": 0.0084,
"1.0": 0.0226
},
"latency_p50_ms": 252.3,
"latency_p95_ms": 390.3,
"cost_usd": 0.00786,
"input_tokens": 187084,
"served": {
"typesafe/jev-1.13-20260917": 399
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.6391,
"kai~jev": 0.9774,
"laya~jev": 0.6516
}
},
"typed_decisions": {
"kai": {
"n": 1635,
"accuracy": 0.756,
"macro_f1": 0.7248,
"ece": 0.1568,
"brier": 0.3808,
"nll": 0.6711,
"mean_confidence": 0.5994,
"acc_at_50_coverage": 0.9009,
"unanswered": 365,
"zero_prob": 0.0,
"questions": 2000,
"aurc": 0.1122,
"risk_at_coverage": {
"0.1": 0.0184,
"0.25": 0.0392,
"0.5": 0.0991,
"0.75": 0.1803,
"0.9": 0.2175,
"1.0": 0.244
},
"soft_accuracy": 0.4887,
"brier_vs_soft": 0.068,
"score_mae": 0.2484,
"within_1_level": 0.9847,
"by_workflow": {
"agent_trace_observability": 0.7367,
"customer_service": 0.6849,
"invoice_processing": 0.8438,
"security_incidents": 0.7639
},
"by_question_type": {
"choice": 0.7088,
"noul": 0.8385,
"score": 0.7309
}
},
"laya": {
"n": 2000,
"accuracy": 0.3615,
"macro_f1": 0.3266,
"ece": 0.1747,
"brier": 0.7497,
"nll": 1.3227,
"mean_confidence": 0.5362,
"acc_at_50_coverage": 0.428,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 2000,
"aurc": 0.5929,
"risk_at_coverage": {
"0.1": 0.605,
"0.25": 0.604,
"0.5": 0.572,
"0.75": 0.6033,
"0.9": 0.6239,
"1.0": 0.6385
},
"soft_accuracy": 0.3315,
"brier_vs_soft": 0.3155,
"score_mae": 0.6937,
"within_1_level": 0.7525,
"by_workflow": {
"agent_trace_observability": 0.388,
"customer_service": 0.382,
"invoice_processing": 0.36,
"security_incidents": 0.316
},
"by_question_type": {
"choice": 0.2883,
"noul": 0.4867,
"score": 0.3225
}
},
"laya:english": {
"n": 2000,
"accuracy": 0.3615,
"macro_f1": 0.3266,
"ece": 0.1747,
"brier": 0.7497,
"nll": 1.3227,
"mean_confidence": 0.5362,
"acc_at_50_coverage": 0.428,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 2000,
"aurc": 0.5929,
"risk_at_coverage": {
"0.1": 0.605,
"0.25": 0.604,
"0.5": 0.572,
"0.75": 0.6033,
"0.9": 0.6239,
"1.0": 0.6385
},
"soft_accuracy": 0.3315,
"brier_vs_soft": 0.3155,
"score_mae": 0.6937,
"within_1_level": 0.7525,
"by_workflow": {
"agent_trace_observability": 0.388,
"customer_service": 0.382,
"invoice_processing": 0.36,
"security_incidents": 0.316
},
"by_question_type": {
"choice": 0.2883,
"noul": 0.4867,
"score": 0.3225
}
},
"laya:multilingual": {
"n": 2000,
"accuracy": 0.352,
"macro_f1": 0.3503,
"ece": 0.3139,
"brier": 0.8904,
"nll": 1.8243,
"mean_confidence": 0.6629,
"acc_at_50_coverage": 0.414,
"unanswered": 0,
"zero_prob": 0.0005,
"questions": 2000,
"aurc": 0.5563,
"risk_at_coverage": {
"0.1": 0.45,
"0.25": 0.5,
"0.5": 0.586,
"0.75": 0.6353,
"0.9": 0.645,
"1.0": 0.648
},
"soft_accuracy": 0.328,
"brier_vs_soft": 0.4626,
"score_mae": 0.7604,
"within_1_level": 0.7,
"by_workflow": {
"agent_trace_observability": 0.286,
"customer_service": 0.42,
"invoice_processing": 0.308,
"security_incidents": 0.394
},
"by_question_type": {
"choice": 0.295,
"noul": 0.4967,
"score": 0.2863
}
},
"laya:typed-decisions": {
"n": 2000,
"accuracy": 0.766,
"macro_f1": 0.7503,
"ece": 0.2132,
"brier": 0.3996,
"nll": 0.7069,
"mean_confidence": 0.5528,
"acc_at_50_coverage": 0.903,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 2000,
"aurc": 0.0988,
"risk_at_coverage": {
"0.1": 0.005,
"0.25": 0.028,
"0.5": 0.097,
"0.75": 0.1647,
"0.9": 0.2044,
"1.0": 0.234
},
"soft_accuracy": 0.4707,
"brier_vs_soft": 0.0615,
"score_mae": 0.2424,
"within_1_level": 0.995,
"by_workflow": {
"agent_trace_observability": 0.73,
"customer_service": 0.764,
"invoice_processing": 0.804,
"security_incidents": 0.766
},
"by_question_type": {
"choice": 0.7333,
"noul": 0.8567,
"score": 0.7225
}
},
"jev": {
"n": 2000,
"accuracy": 0.7355,
"macro_f1": 0.7381,
"ece": 0.0472,
"brier": 0.3658,
"nll": 0.7067,
"mean_confidence": 0.7567,
"acc_at_50_coverage": 0.885,
"unanswered": 0,
"zero_prob": 0.003,
"questions": 2000,
"aurc": 0.1277,
"risk_at_coverage": {
"0.1": 0.025,
"0.25": 0.0769,
"0.5": 0.1142,
"0.75": 0.1859,
"0.9": 0.2314,
"1.0": 0.2645
},
"soft_accuracy": 0.5383,
"brier_vs_soft": 0.1483,
"score_mae": 0.3912,
"within_1_level": 0.9463,
"by_workflow": {
"agent_trace_observability": 0.634,
"customer_service": 0.79,
"invoice_processing": 0.778,
"security_incidents": 0.74
},
"by_question_type": {
"choice": 0.7317,
"noul": 0.785,
"score": 0.7013
},
"latency_p50_ms": 255.4,
"latency_p95_ms": 340.5,
"cost_usd": 0.01587,
"input_tokens": 377758,
"served": {
"typesafe/jev-1.13-20260917": 400
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.3352,
"kai~jev": 0.7144,
"laya~jev": 0.335
}
},
"massive.af": {
"kai": {
"n": 100,
"accuracy": 0.9,
"macro_f1": 0.8853,
"ece": 0.0581,
"brier": 0.1471,
"nll": 0.3311,
"mean_confidence": 0.9581,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0091,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0444,
"1.0": 0.1
}
},
"laya": {
"n": 100,
"accuracy": 0.34,
"macro_f1": 0.3367,
"ece": 0.4845,
"brier": 1.0683,
"nll": 5.0409,
"mean_confidence": 0.8245,
"acc_at_50_coverage": 0.48,
"unanswered": 0,
"zero_prob": 0.09,
"questions": 100,
"aurc": 0.4721,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.4,
"0.5": 0.52,
"0.75": 0.6,
"0.9": 0.6333,
"1.0": 0.66
}
},
"laya:english": {
"n": 100,
"accuracy": 0.29,
"macro_f1": 0.2874,
"ece": 0.5588,
"brier": 1.1918,
"nll": 5.9153,
"mean_confidence": 0.8488,
"acc_at_50_coverage": 0.42,
"unanswered": 0,
"zero_prob": 0.11,
"questions": 100,
"aurc": 0.5205,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.48,
"0.5": 0.58,
"0.75": 0.6667,
"0.9": 0.6778,
"1.0": 0.71
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.35,
"macro_f1": 0.3411,
"ece": 0.4394,
"brier": 0.9812,
"nll": 3.2326,
"mean_confidence": 0.7894,
"acc_at_50_coverage": 0.58,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.4303,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.32,
"0.5": 0.42,
"0.75": 0.56,
"0.9": 0.6111,
"1.0": 0.65
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.32,
"macro_f1": 0.2902,
"ece": 0.4265,
"brier": 1.0065,
"nll": 3.4483,
"mean_confidence": 0.7344,
"acc_at_50_coverage": 0.52,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.5088,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.4,
"0.5": 0.48,
"0.75": 0.6,
"0.9": 0.6444,
"1.0": 0.68
}
},
"jev": {
"n": 100,
"accuracy": 0.87,
"macro_f1": 0.8625,
"ece": 0.0873,
"brier": 0.2058,
"nll": 0.934,
"mean_confidence": 0.8823,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0259,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.04,
"0.9": 0.0667,
"1.0": 0.13
},
"latency_p50_ms": 249.9,
"latency_p95_ms": 342.1,
"cost_usd": 0.00262,
"input_tokens": 62358,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.34,
"kai~jev": 0.82,
"laya~jev": 0.33
}
},
"massive.am": {
"kai": {
"n": 100,
"accuracy": 0.86,
"macro_f1": 0.8565,
"ece": 0.0849,
"brier": 0.221,
"nll": 0.5804,
"mean_confidence": 0.9245,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0209,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0889,
"1.0": 0.14
}
},
"laya": {
"n": 100,
"accuracy": 0.15,
"macro_f1": 0.1307,
"ece": 0.4839,
"brier": 1.2188,
"nll": 4.266,
"mean_confidence": 0.6339,
"acc_at_50_coverage": 0.26,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.8162,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 0.76,
"0.5": 0.74,
"0.75": 0.8267,
"0.9": 0.8444,
"1.0": 0.85
}
},
"laya:english": {
"n": 100,
"accuracy": 0.12,
"macro_f1": 0.1122,
"ece": 0.6804,
"brier": 1.461,
"nll": 8.0305,
"mean_confidence": 0.7873,
"acc_at_50_coverage": 0.12,
"unanswered": 0,
"zero_prob": 0.07,
"questions": 100,
"aurc": 0.8853,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.88,
"0.5": 0.88,
"0.75": 0.8533,
"0.9": 0.8778,
"1.0": 0.88
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.15,
"macro_f1": 0.1307,
"ece": 0.4839,
"brier": 1.2188,
"nll": 4.266,
"mean_confidence": 0.6339,
"acc_at_50_coverage": 0.26,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.8162,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 0.76,
"0.5": 0.74,
"0.75": 0.8267,
"0.9": 0.8444,
"1.0": 0.85
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.16,
"macro_f1": 0.1328,
"ece": 0.4891,
"brier": 1.2868,
"nll": 4.7541,
"mean_confidence": 0.6439,
"acc_at_50_coverage": 0.12,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8667,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.88,
"0.5": 0.88,
"0.75": 0.8267,
"0.9": 0.8333,
"1.0": 0.84
}
},
"jev": {
"n": 100,
"accuracy": 0.85,
"macro_f1": 0.8632,
"ece": 0.096,
"brier": 0.2195,
"nll": 1.0045,
"mean_confidence": 0.8305,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0284,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.02,
"0.75": 0.04,
"0.9": 0.0778,
"1.0": 0.15
},
"latency_p50_ms": 220.4,
"latency_p95_ms": 321.1,
"cost_usd": 0.00273,
"input_tokens": 64888,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.15,
"kai~jev": 0.81,
"laya~jev": 0.19
}
},
"massive.ar": {
"kai": {
"n": 100,
"accuracy": 0.91,
"macro_f1": 0.8931,
"ece": 0.073,
"brier": 0.1227,
"nll": 0.3184,
"mean_confidence": 0.9304,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0075,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0333,
"1.0": 0.09
}
},
"laya": {
"n": 100,
"accuracy": 0.46,
"macro_f1": 0.4325,
"ece": 0.315,
"brier": 0.8464,
"nll": 2.7141,
"mean_confidence": 0.775,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.3768,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.24,
"0.5": 0.4,
"0.75": 0.4533,
"0.9": 0.5,
"1.0": 0.54
}
},
"laya:english": {
"n": 100,
"accuracy": 0.11,
"macro_f1": 0.093,
"ece": 0.5795,
"brier": 1.3826,
"nll": 6.5867,
"mean_confidence": 0.6833,
"acc_at_50_coverage": 0.12,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.8012,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.8,
"0.5": 0.88,
"0.75": 0.88,
"0.9": 0.9,
"1.0": 0.89
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.46,
"macro_f1": 0.4325,
"ece": 0.315,
"brier": 0.8464,
"nll": 2.7141,
"mean_confidence": 0.775,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.3901,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.24,
"0.5": 0.4,
"0.75": 0.4533,
"0.9": 0.5,
"1.0": 0.54
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.08,
"macro_f1": 0.0617,
"ece": 0.4425,
"brier": 1.1943,
"nll": 4.4448,
"mean_confidence": 0.5225,
"acc_at_50_coverage": 0.14,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8303,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.8,
"0.5": 0.86,
"0.75": 0.8933,
"0.9": 0.9111,
"1.0": 0.92
}
},
"jev": {
"n": 100,
"accuracy": 0.87,
"macro_f1": 0.8675,
"ece": 0.0756,
"brier": 0.1962,
"nll": 0.9165,
"mean_confidence": 0.8494,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0202,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0778,
"1.0": 0.13
},
"latency_p50_ms": 240.2,
"latency_p95_ms": 353.2,
"cost_usd": 0.00264,
"input_tokens": 62814,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.46,
"kai~jev": 0.82,
"laya~jev": 0.44
}
},
"massive.az": {
"kai": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9386,
"ece": 0.0377,
"brier": 0.0772,
"nll": 0.2089,
"mean_confidence": 0.9544,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0036,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0111,
"1.0": 0.06
}
},
"laya": {
"n": 100,
"accuracy": 0.36,
"macro_f1": 0.3361,
"ece": 0.3929,
"brier": 0.9729,
"nll": 3.3402,
"mean_confidence": 0.733,
"acc_at_50_coverage": 0.52,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.4833,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.4,
"0.5": 0.48,
"0.75": 0.6,
"0.9": 0.6222,
"1.0": 0.64
}
},
"laya:english": {
"n": 100,
"accuracy": 0.1,
"macro_f1": 0.0629,
"ece": 0.5983,
"brier": 1.4097,
"nll": 6.6948,
"mean_confidence": 0.6983,
"acc_at_50_coverage": 0.14,
"unanswered": 0,
"zero_prob": 0.05,
"questions": 100,
"aurc": 0.7809,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.72,
"0.5": 0.86,
"0.75": 0.8933,
"0.9": 0.9,
"1.0": 0.9
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.36,
"macro_f1": 0.3361,
"ece": 0.381,
"brier": 0.9578,
"nll": 3.1981,
"mean_confidence": 0.7211,
"acc_at_50_coverage": 0.54,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.4695,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.36,
"0.5": 0.46,
"0.75": 0.5867,
"0.9": 0.6222,
"1.0": 0.64
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.07,
"macro_f1": 0.0464,
"ece": 0.4652,
"brier": 1.2345,
"nll": 4.9386,
"mean_confidence": 0.5352,
"acc_at_50_coverage": 0.1,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8464,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.8,
"0.5": 0.9,
"0.75": 0.9067,
"0.9": 0.9222,
"1.0": 0.93
}
},
"jev": {
"n": 100,
"accuracy": 0.88,
"macro_f1": 0.8743,
"ece": 0.0482,
"brier": 0.1693,
"nll": 1.1198,
"mean_confidence": 0.8946,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 0.0135,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0556,
"1.0": 0.12
},
"latency_p50_ms": 245.1,
"latency_p95_ms": 383.1,
"cost_usd": 0.00264,
"input_tokens": 62750,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.38,
"kai~jev": 0.86,
"laya~jev": 0.38
}
},
"massive.bn": {
"kai": {
"n": 100,
"accuracy": 0.91,
"macro_f1": 0.8963,
"ece": 0.0613,
"brier": 0.144,
"nll": 0.5073,
"mean_confidence": 0.9399,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0126,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0333,
"1.0": 0.09
}
},
"laya": {
"n": 100,
"accuracy": 0.45,
"macro_f1": 0.44,
"ece": 0.354,
"brier": 0.8976,
"nll": 2.868,
"mean_confidence": 0.7788,
"acc_at_50_coverage": 0.54,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.4341,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.4,
"0.5": 0.46,
"0.75": 0.5067,
"0.9": 0.5222,
"1.0": 0.55
}
},
"laya:english": {
"n": 100,
"accuracy": 0.08,
"macro_f1": 0.056,
"ece": 0.6267,
"brier": 1.4117,
"nll": 6.4245,
"mean_confidence": 0.7067,
"acc_at_50_coverage": 0.12,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.8268,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.84,
"0.5": 0.88,
"0.75": 0.8933,
"0.9": 0.9111,
"1.0": 0.92
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.45,
"macro_f1": 0.44,
"ece": 0.354,
"brier": 0.8976,
"nll": 2.868,
"mean_confidence": 0.7788,
"acc_at_50_coverage": 0.54,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.4341,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.4,
"0.5": 0.46,
"0.75": 0.5067,
"0.9": 0.5222,
"1.0": 0.55
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.12,
"macro_f1": 0.1157,
"ece": 0.4204,
"brier": 1.2058,
"nll": 4.4151,
"mean_confidence": 0.5404,
"acc_at_50_coverage": 0.12,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8005,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.8,
"0.5": 0.88,
"0.75": 0.88,
"0.9": 0.8889,
"1.0": 0.88
}
},
"jev": {
"n": 100,
"accuracy": 0.86,
"macro_f1": 0.8345,
"ece": 0.0672,
"brier": 0.1895,
"nll": 0.6325,
"mean_confidence": 0.8897,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0265,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0944,
"1.0": 0.14
},
"latency_p50_ms": 249.0,
"latency_p95_ms": 349.8,
"cost_usd": 0.00267,
"input_tokens": 63492,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.43,
"kai~jev": 0.84,
"laya~jev": 0.48
}
},
"massive.cy": {
"kai": {
"n": 100,
"accuracy": 0.86,
"macro_f1": 0.8568,
"ece": 0.0689,
"brier": 0.1931,
"nll": 0.5072,
"mean_confidence": 0.9241,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0236,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.02,
"0.75": 0.0133,
"0.9": 0.0778,
"1.0": 0.14
}
},
"laya": {
"n": 100,
"accuracy": 0.16,
"macro_f1": 0.1371,
"ece": 0.6107,
"brier": 1.3894,
"nll": 6.9922,
"mean_confidence": 0.7707,
"acc_at_50_coverage": 0.2,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.7641,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.8,
"0.5": 0.8,
"0.75": 0.8267,
"0.9": 0.8222,
"1.0": 0.84
}
},
"laya:english": {
"n": 100,
"accuracy": 0.12,
"macro_f1": 0.0906,
"ece": 0.6452,
"brier": 1.4675,
"nll": 7.4227,
"mean_confidence": 0.7652,
"acc_at_50_coverage": 0.14,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.8194,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.84,
"0.5": 0.86,
"0.75": 0.8667,
"0.9": 0.8667,
"1.0": 0.88
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.13,
"macro_f1": 0.1217,
"ece": 0.5873,
"brier": 1.3123,
"nll": 4.9721,
"mean_confidence": 0.7173,
"acc_at_50_coverage": 0.22,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 0.7637,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.72,
"0.5": 0.78,
"0.75": 0.8267,
"0.9": 0.8556,
"1.0": 0.87
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.17,
"macro_f1": 0.1322,
"ece": 0.4455,
"brier": 1.2037,
"nll": 4.8677,
"mean_confidence": 0.6155,
"acc_at_50_coverage": 0.22,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.6951,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.68,
"0.5": 0.78,
"0.75": 0.8,
"0.9": 0.8222,
"1.0": 0.83
}
},
"jev": {
"n": 100,
"accuracy": 0.76,
"macro_f1": 0.7637,
"ece": 0.0878,
"brier": 0.3358,
"nll": 1.7761,
"mean_confidence": 0.7714,
"acc_at_50_coverage": 0.96,
"unanswered": 0,
"zero_prob": 0.04,
"questions": 100,
"aurc": 0.0565,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.04,
"0.75": 0.0667,
"0.9": 0.1556,
"1.0": 0.24
},
"latency_p50_ms": 255.3,
"latency_p95_ms": 329.0,
"cost_usd": 0.00263,
"input_tokens": 62589,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.19,
"kai~jev": 0.77,
"laya~jev": 0.18
}
},
"massive.da": {
"kai": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.9094,
"ece": 0.0575,
"brier": 0.1205,
"nll": 0.3333,
"mean_confidence": 0.967,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0062,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0222,
"1.0": 0.08
}
},
"laya": {
"n": 100,
"accuracy": 0.46,
"macro_f1": 0.4167,
"ece": 0.4518,
"brier": 0.934,
"nll": 4.4891,
"mean_confidence": 0.8924,
"acc_at_50_coverage": 0.62,
"unanswered": 0,
"zero_prob": 0.05,
"questions": 100,
"aurc": 0.3267,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.24,
"0.5": 0.38,
"0.75": 0.4533,
"0.9": 0.5111,
"1.0": 0.54
}
},
"laya:english": {
"n": 100,
"accuracy": 0.35,
"macro_f1": 0.3082,
"ece": 0.526,
"brier": 1.1294,
"nll": 6.2405,
"mean_confidence": 0.876,
"acc_at_50_coverage": 0.52,
"unanswered": 0,
"zero_prob": 0.07,
"questions": 100,
"aurc": 0.3864,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.2,
"0.5": 0.48,
"0.75": 0.5733,
"0.9": 0.6222,
"1.0": 0.65
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.52,
"macro_f1": 0.4956,
"ece": 0.315,
"brier": 0.7303,
"nll": 2.1047,
"mean_confidence": 0.8062,
"acc_at_50_coverage": 0.7,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.269,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.12,
"0.5": 0.3,
"0.75": 0.3867,
"0.9": 0.4444,
"1.0": 0.48
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.34,
"macro_f1": 0.3197,
"ece": 0.3517,
"brier": 0.9178,
"nll": 3.487,
"mean_confidence": 0.6917,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.3578,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.08,
"0.5": 0.4,
"0.75": 0.5467,
"0.9": 0.6222,
"1.0": 0.66
}
},
"jev": {
"n": 100,
"accuracy": 0.88,
"macro_f1": 0.8633,
"ece": 0.0439,
"brier": 0.1538,
"nll": 0.8107,
"mean_confidence": 0.8905,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0153,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0556,
"1.0": 0.12
},
"latency_p50_ms": 243.2,
"latency_p95_ms": 371.7,
"cost_usd": 0.00262,
"input_tokens": 62279,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.47,
"kai~jev": 0.85,
"laya~jev": 0.47
}
},
"massive.de": {
"kai": {
"n": 100,
"accuracy": 0.91,
"macro_f1": 0.9032,
"ece": 0.0514,
"brier": 0.1135,
"nll": 0.3163,
"mean_confidence": 0.9449,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0066,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0111,
"1.0": 0.09
}
},
"laya": {
"n": 100,
"accuracy": 0.53,
"macro_f1": 0.5045,
"ece": 0.3037,
"brier": 0.7303,
"nll": 2.8055,
"mean_confidence": 0.831,
"acc_at_50_coverage": 0.76,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 0.2334,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.08,
"0.5": 0.24,
"0.75": 0.36,
"0.9": 0.4333,
"1.0": 0.47
}
},
"laya:english": {
"n": 100,
"accuracy": 0.42,
"macro_f1": 0.387,
"ece": 0.452,
"brier": 1.0025,
"nll": 5.0184,
"mean_confidence": 0.872,
"acc_at_50_coverage": 0.68,
"unanswered": 0,
"zero_prob": 0.08,
"questions": 100,
"aurc": 0.3333,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.12,
"0.5": 0.32,
"0.75": 0.5067,
"0.9": 0.5667,
"1.0": 0.58
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.5,
"macro_f1": 0.4768,
"ece": 0.3358,
"brier": 0.7588,
"nll": 2.3681,
"mean_confidence": 0.8131,
"acc_at_50_coverage": 0.76,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.2531,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.08,
"0.5": 0.24,
"0.75": 0.3867,
"0.9": 0.4667,
"1.0": 0.5
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.39,
"macro_f1": 0.3745,
"ece": 0.3393,
"brier": 0.8496,
"nll": 2.6918,
"mean_confidence": 0.7293,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.362,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.24,
"0.5": 0.4,
"0.75": 0.52,
"0.9": 0.5667,
"1.0": 0.61
}
},
"jev": {
"n": 99,
"accuracy": 0.9394,
"macro_f1": 0.9387,
"ece": 0.057,
"brier": 0.087,
"nll": 0.4212,
"mean_confidence": 0.935,
"acc_at_50_coverage": 1.0,
"unanswered": 1,
"zero_prob": 0.0101,
"questions": 100,
"aurc": 0.0073,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0135,
"0.9": 0.0225,
"1.0": 0.0606
},
"latency_p50_ms": 255.2,
"latency_p95_ms": 340.8,
"cost_usd": 0.00259,
"input_tokens": 61688,
"served": {
"typesafe/jev-1.13-20260917": 99
},
"n_errors": 1
},
"agreement": {
"kai~laya": 0.54,
"kai~jev": 0.8687,
"laya~jev": 0.5051
}
},
"massive.el": {
"kai": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9259,
"ece": 0.0632,
"brier": 0.0984,
"nll": 0.2456,
"mean_confidence": 0.9423,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0054,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0222,
"1.0": 0.06
}
},
"laya": {
"n": 100,
"accuracy": 0.44,
"macro_f1": 0.4148,
"ece": 0.4297,
"brier": 0.9225,
"nll": 3.2259,
"mean_confidence": 0.801,
"acc_at_50_coverage": 0.58,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.4247,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.4,
"0.5": 0.42,
"0.75": 0.52,
"0.9": 0.5444,
"1.0": 0.56
}
},
"laya:english": {
"n": 100,
"accuracy": 0.13,
"macro_f1": 0.1,
"ece": 0.6789,
"brier": 1.4564,
"nll": 8.1375,
"mean_confidence": 0.8089,
"acc_at_50_coverage": 0.2,
"unanswered": 0,
"zero_prob": 0.13,
"questions": 100,
"aurc": 0.7909,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.76,
"0.5": 0.8,
"0.75": 0.84,
"0.9": 0.8556,
"1.0": 0.87
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.44,
"macro_f1": 0.4148,
"ece": 0.4297,
"brier": 0.9225,
"nll": 3.2259,
"mean_confidence": 0.801,
"acc_at_50_coverage": 0.58,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.4247,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.4,
"0.5": 0.42,
"0.75": 0.52,
"0.9": 0.5444,
"1.0": 0.56
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.13,
"macro_f1": 0.0946,
"ece": 0.4772,
"brier": 1.2128,
"nll": 5.069,
"mean_confidence": 0.6072,
"acc_at_50_coverage": 0.2,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.7676,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.68,
"0.5": 0.8,
"0.75": 0.84,
"0.9": 0.8667,
"1.0": 0.87
}
},
"jev": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9413,
"ece": 0.0571,
"brier": 0.094,
"nll": 0.4307,
"mean_confidence": 0.9197,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0085,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0222,
"1.0": 0.06
},
"latency_p50_ms": 244.2,
"latency_p95_ms": 337.1,
"cost_usd": 0.00267,
"input_tokens": 63560,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.44,
"kai~jev": 0.9,
"laya~jev": 0.45
}
},
"massive.en": {
"kai": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9334,
"ece": 0.0513,
"brier": 0.1065,
"nll": 0.312,
"mean_confidence": 0.9835,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0054,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0222,
"1.0": 0.06
}
},
"laya": {
"n": 100,
"accuracy": 0.82,
"macro_f1": 0.7876,
"ece": 0.1382,
"brier": 0.2882,
"nll": 1.2282,
"mean_confidence": 0.9582,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0386,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.02,
"0.75": 0.0533,
"0.9": 0.1111,
"1.0": 0.18
}
},
"laya:english": {
"n": 100,
"accuracy": 0.82,
"macro_f1": 0.7876,
"ece": 0.1382,
"brier": 0.2882,
"nll": 1.2282,
"mean_confidence": 0.9582,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0386,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.02,
"0.75": 0.0533,
"0.9": 0.1111,
"1.0": 0.18
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.71,
"macro_f1": 0.6689,
"ece": 0.2333,
"brier": 0.5062,
"nll": 1.639,
"mean_confidence": 0.8876,
"acc_at_50_coverage": 0.92,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.1252,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.08,
"0.5": 0.08,
"0.75": 0.2,
"0.9": 0.2556,
"1.0": 0.29
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.82,
"macro_f1": 0.7839,
"ece": 0.0932,
"brier": 0.2756,
"nll": 0.7103,
"mean_confidence": 0.9006,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0405,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.02,
"0.75": 0.0533,
"0.9": 0.1222,
"1.0": 0.18
}
},
"jev": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.9093,
"ece": 0.0253,
"brier": 0.108,
"nll": 0.6887,
"mean_confidence": 0.9413,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.008,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0117,
"0.9": 0.0222,
"1.0": 0.08
},
"latency_p50_ms": 255.6,
"latency_p95_ms": 344.2,
"cost_usd": 0.00261,
"input_tokens": 62052,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.8,
"kai~jev": 0.9,
"laya~jev": 0.81
}
},
"massive.es": {
"kai": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.91,
"ece": 0.0672,
"brier": 0.1325,
"nll": 0.3483,
"mean_confidence": 0.9534,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0129,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0444,
"1.0": 0.08
}
},
"laya": {
"n": 100,
"accuracy": 0.53,
"macro_f1": 0.5044,
"ece": 0.3476,
"brier": 0.8103,
"nll": 3.748,
"mean_confidence": 0.8509,
"acc_at_50_coverage": 0.62,
"unanswered": 0,
"zero_prob": 0.09,
"questions": 100,
"aurc": 0.3585,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.36,
"0.5": 0.38,
"0.75": 0.4267,
"0.9": 0.4556,
"1.0": 0.47
}
},
"laya:english": {
"n": 100,
"accuracy": 0.51,
"macro_f1": 0.4767,
"ece": 0.4225,
"brier": 0.8845,
"nll": 5.0115,
"mean_confidence": 0.9245,
"acc_at_50_coverage": 0.62,
"unanswered": 0,
"zero_prob": 0.16,
"questions": 100,
"aurc": 0.3235,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.24,
"0.5": 0.38,
"0.75": 0.4133,
"0.9": 0.4556,
"1.0": 0.49
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.58,
"macro_f1": 0.5213,
"ece": 0.2577,
"brier": 0.6722,
"nll": 1.9703,
"mean_confidence": 0.796,
"acc_at_50_coverage": 0.7,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.2623,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.2,
"0.5": 0.3,
"0.75": 0.36,
"0.9": 0.4,
"1.0": 0.42
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.45,
"macro_f1": 0.4176,
"ece": 0.2835,
"brier": 0.7533,
"nll": 2.5014,
"mean_confidence": 0.723,
"acc_at_50_coverage": 0.7,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.2888,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.12,
"0.5": 0.3,
"0.75": 0.48,
"0.9": 0.5222,
"1.0": 0.55
}
},
"jev": {
"n": 100,
"accuracy": 0.93,
"macro_f1": 0.9326,
"ece": 0.0628,
"brier": 0.1223,
"nll": 0.4934,
"mean_confidence": 0.9297,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0132,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0444,
"1.0": 0.07
},
"latency_p50_ms": 240.1,
"latency_p95_ms": 361.0,
"cost_usd": 0.00261,
"input_tokens": 62204,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.54,
"kai~jev": 0.9,
"laya~jev": 0.52
}
},
"massive.fa": {
"kai": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9137,
"ece": 0.0382,
"brier": 0.089,
"nll": 0.2553,
"mean_confidence": 0.9715,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0038,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0111,
"1.0": 0.06
}
},
"laya": {
"n": 100,
"accuracy": 0.51,
"macro_f1": 0.4728,
"ece": 0.2985,
"brier": 0.7935,
"nll": 2.6907,
"mean_confidence": 0.7957,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.3877,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.28,
"0.5": 0.4,
"0.75": 0.4267,
"0.9": 0.4333,
"1.0": 0.49
}
},
"laya:english": {
"n": 100,
"accuracy": 0.14,
"macro_f1": 0.1197,
"ece": 0.6331,
"brier": 1.453,
"nll": 7.3943,
"mean_confidence": 0.7731,
"acc_at_50_coverage": 0.16,
"unanswered": 0,
"zero_prob": 0.09,
"questions": 100,
"aurc": 0.8445,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.84,
"0.5": 0.84,
"0.75": 0.8533,
"0.9": 0.8444,
"1.0": 0.86
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.51,
"macro_f1": 0.4728,
"ece": 0.2985,
"brier": 0.7935,
"nll": 2.6907,
"mean_confidence": 0.7957,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.3877,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.28,
"0.5": 0.4,
"0.75": 0.4267,
"0.9": 0.4333,
"1.0": 0.49
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.09,
"macro_f1": 0.0856,
"ece": 0.4465,
"brier": 1.2247,
"nll": 4.5383,
"mean_confidence": 0.5365,
"acc_at_50_coverage": 0.12,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.9094,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.92,
"0.5": 0.88,
"0.75": 0.9067,
"0.9": 0.9,
"1.0": 0.91
}
},
"jev": {
"n": 100,
"accuracy": 0.95,
"macro_f1": 0.9524,
"ece": 0.0535,
"brier": 0.0779,
"nll": 0.4136,
"mean_confidence": 0.9294,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0059,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0111,
"1.0": 0.05
},
"latency_p50_ms": 227.3,
"latency_p95_ms": 330.3,
"cost_usd": 0.00265,
"input_tokens": 63079,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.51,
"kai~jev": 0.9,
"laya~jev": 0.5
}
},
"massive.fi": {
"kai": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.9112,
"ece": 0.0562,
"brier": 0.1126,
"nll": 0.3371,
"mean_confidence": 0.9655,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0186,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.04,
"0.5": 0.02,
"0.75": 0.0133,
"0.9": 0.0222,
"1.0": 0.08
}
},
"laya": {
"n": 100,
"accuracy": 0.3,
"macro_f1": 0.2993,
"ece": 0.4618,
"brier": 1.1197,
"nll": 4.6211,
"mean_confidence": 0.7592,
"acc_at_50_coverage": 0.38,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.5732,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.48,
"0.5": 0.62,
"0.75": 0.6667,
"0.9": 0.6778,
"1.0": 0.7
}
},
"laya:english": {
"n": 100,
"accuracy": 0.13,
"macro_f1": 0.1213,
"ece": 0.7189,
"brier": 1.5179,
"nll": 8.0296,
"mean_confidence": 0.8489,
"acc_at_50_coverage": 0.16,
"unanswered": 0,
"zero_prob": 0.13,
"questions": 100,
"aurc": 0.7984,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.72,
"0.5": 0.84,
"0.75": 0.84,
"0.9": 0.8556,
"1.0": 0.87
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.34,
"macro_f1": 0.3282,
"ece": 0.3918,
"brier": 1.0128,
"nll": 3.3753,
"mean_confidence": 0.7316,
"acc_at_50_coverage": 0.46,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.5655,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.48,
"0.5": 0.54,
"0.75": 0.6267,
"0.9": 0.6333,
"1.0": 0.66
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.12,
"macro_f1": 0.1091,
"ece": 0.515,
"brier": 1.2553,
"nll": 4.6869,
"mean_confidence": 0.635,
"acc_at_50_coverage": 0.2,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.7582,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.76,
"0.5": 0.8,
"0.75": 0.8533,
"0.9": 0.8667,
"1.0": 0.88
}
},
"jev": {
"n": 100,
"accuracy": 0.89,
"macro_f1": 0.8792,
"ece": 0.0383,
"brier": 0.1513,
"nll": 0.7839,
"mean_confidence": 0.8981,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.016,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0667,
"1.0": 0.11
},
"latency_p50_ms": 229.7,
"latency_p95_ms": 355.8,
"cost_usd": 0.00262,
"input_tokens": 62493,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.3,
"kai~jev": 0.87,
"laya~jev": 0.28
}
},
"massive.fr": {
"kai": {
"n": 100,
"accuracy": 0.91,
"macro_f1": 0.8957,
"ece": 0.0745,
"brier": 0.1401,
"nll": 0.3909,
"mean_confidence": 0.9663,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0107,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0333,
"1.0": 0.09
}
},
"laya": {
"n": 100,
"accuracy": 0.6,
"macro_f1": 0.5817,
"ece": 0.2683,
"brier": 0.6627,
"nll": 2.6707,
"mean_confidence": 0.8683,
"acc_at_50_coverage": 0.76,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.2329,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.24,
"0.5": 0.24,
"0.75": 0.3067,
"0.9": 0.3444,
"1.0": 0.4
}
},
"laya:english": {
"n": 100,
"accuracy": 0.59,
"macro_f1": 0.5822,
"ece": 0.2995,
"brier": 0.694,
"nll": 3.2748,
"mean_confidence": 0.8895,
"acc_at_50_coverage": 0.78,
"unanswered": 0,
"zero_prob": 0.05,
"questions": 100,
"aurc": 0.2064,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.08,
"0.5": 0.22,
"0.75": 0.3067,
"0.9": 0.3667,
"1.0": 0.41
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.6,
"macro_f1": 0.5733,
"ece": 0.2484,
"brier": 0.6184,
"nll": 2.2053,
"mean_confidence": 0.8403,
"acc_at_50_coverage": 0.8,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.2022,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.16,
"0.5": 0.2,
"0.75": 0.2667,
"0.9": 0.3333,
"1.0": 0.4
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.6,
"macro_f1": 0.5869,
"ece": 0.228,
"brier": 0.638,
"nll": 1.8988,
"mean_confidence": 0.7941,
"acc_at_50_coverage": 0.78,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.1975,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.04,
"0.5": 0.22,
"0.75": 0.3467,
"0.9": 0.3667,
"1.0": 0.4
}
},
"jev": {
"n": 100,
"accuracy": 0.9,
"macro_f1": 0.896,
"ece": 0.0652,
"brier": 0.1302,
"nll": 0.4976,
"mean_confidence": 0.9144,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0117,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0444,
"1.0": 0.1
},
"latency_p50_ms": 223.5,
"latency_p95_ms": 319.3,
"cost_usd": 0.00262,
"input_tokens": 62358,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.63,
"kai~jev": 0.86,
"laya~jev": 0.6
}
},
"massive.he": {
"kai": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.9077,
"ece": 0.0593,
"brier": 0.1452,
"nll": 0.4688,
"mean_confidence": 0.958,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0138,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0444,
"1.0": 0.08
}
},
"laya": {
"n": 100,
"accuracy": 0.37,
"macro_f1": 0.3581,
"ece": 0.4186,
"brier": 0.9688,
"nll": 2.8349,
"mean_confidence": 0.787,
"acc_at_50_coverage": 0.58,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.4009,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.28,
"0.5": 0.42,
"0.75": 0.56,
"0.9": 0.6,
"1.0": 0.63
}
},
"laya:english": {
"n": 100,
"accuracy": 0.06,
"macro_f1": 0.039,
"ece": 0.7379,
"brier": 1.6073,
"nll": 8.523,
"mean_confidence": 0.7842,
"acc_at_50_coverage": 0.02,
"unanswered": 0,
"zero_prob": 0.11,
"questions": 100,
"aurc": 0.9793,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 1.0,
"0.5": 0.98,
"0.75": 0.96,
"0.9": 0.9556,
"1.0": 0.94
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.37,
"macro_f1": 0.3581,
"ece": 0.4186,
"brier": 0.9688,
"nll": 2.8349,
"mean_confidence": 0.787,
"acc_at_50_coverage": 0.58,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.4009,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.28,
"0.5": 0.42,
"0.75": 0.56,
"0.9": 0.6,
"1.0": 0.63
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.03,
"macro_f1": 0.0261,
"ece": 0.5053,
"brier": 1.2889,
"nll": 5.2056,
"mean_confidence": 0.5353,
"acc_at_50_coverage": 0.04,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.9626,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.96,
"0.5": 0.96,
"0.75": 0.96,
"0.9": 0.9667,
"1.0": 0.97
}
},
"jev": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9349,
"ece": 0.0427,
"brier": 0.0928,
"nll": 0.4303,
"mean_confidence": 0.9287,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0069,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0222,
"1.0": 0.06
},
"latency_p50_ms": 234.8,
"latency_p95_ms": 357.0,
"cost_usd": 0.00266,
"input_tokens": 63273,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.36,
"kai~jev": 0.92,
"laya~jev": 0.36
}
},
"massive.hi": {
"kai": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.9174,
"ece": 0.0777,
"brier": 0.1276,
"nll": 0.3717,
"mean_confidence": 0.9535,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0068,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0222,
"1.0": 0.08
}
},
"laya": {
"n": 100,
"accuracy": 0.46,
"macro_f1": 0.409,
"ece": 0.3685,
"brier": 0.8783,
"nll": 2.8703,
"mean_confidence": 0.8175,
"acc_at_50_coverage": 0.58,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.3879,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.32,
"0.5": 0.42,
"0.75": 0.48,
"0.9": 0.5111,
"1.0": 0.54
}
},
"laya:english": {
"n": 100,
"accuracy": 0.1,
"macro_f1": 0.0758,
"ece": 0.6416,
"brier": 1.4552,
"nll": 7.4333,
"mean_confidence": 0.7416,
"acc_at_50_coverage": 0.08,
"unanswered": 0,
"zero_prob": 0.08,
"questions": 100,
"aurc": 0.8837,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.88,
"0.5": 0.92,
"0.75": 0.88,
"0.9": 0.8889,
"1.0": 0.9
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.46,
"macro_f1": 0.409,
"ece": 0.3685,
"brier": 0.8783,
"nll": 2.8703,
"mean_confidence": 0.8175,
"acc_at_50_coverage": 0.58,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.3879,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.32,
"0.5": 0.42,
"0.75": 0.48,
"0.9": 0.5111,
"1.0": 0.54
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.11,
"macro_f1": 0.087,
"ece": 0.46,
"brier": 1.2697,
"nll": 5.0621,
"mean_confidence": 0.57,
"acc_at_50_coverage": 0.1,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8761,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.88,
"0.5": 0.9,
"0.75": 0.8933,
"0.9": 0.8889,
"1.0": 0.89
}
},
"jev": {
"n": 100,
"accuracy": 0.95,
"macro_f1": 0.949,
"ece": 0.029,
"brier": 0.0827,
"nll": 0.4076,
"mean_confidence": 0.942,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0068,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0222,
"1.0": 0.05
},
"latency_p50_ms": 229.5,
"latency_p95_ms": 336.1,
"cost_usd": 0.00266,
"input_tokens": 63332,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.46,
"kai~jev": 0.92,
"laya~jev": 0.47
}
},
"massive.hu": {
"kai": {
"n": 100,
"accuracy": 0.9,
"macro_f1": 0.8779,
"ece": 0.0804,
"brier": 0.1619,
"nll": 0.517,
"mean_confidence": 0.9377,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0152,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0556,
"1.0": 0.1
}
},
"laya": {
"n": 100,
"accuracy": 0.34,
"macro_f1": 0.3138,
"ece": 0.4374,
"brier": 1.0516,
"nll": 3.7979,
"mean_confidence": 0.7625,
"acc_at_50_coverage": 0.44,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.5046,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.4,
"0.5": 0.56,
"0.75": 0.6533,
"0.9": 0.6667,
"1.0": 0.66
}
},
"laya:english": {
"n": 100,
"accuracy": 0.09,
"macro_f1": 0.0813,
"ece": 0.6783,
"brier": 1.5325,
"nll": 8.7347,
"mean_confidence": 0.7683,
"acc_at_50_coverage": 0.06,
"unanswered": 0,
"zero_prob": 0.14,
"questions": 100,
"aurc": 0.9155,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.92,
"0.5": 0.94,
"0.75": 0.92,
"0.9": 0.9,
"1.0": 0.91
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.36,
"macro_f1": 0.3339,
"ece": 0.4005,
"brier": 1.0048,
"nll": 3.3634,
"mean_confidence": 0.7457,
"acc_at_50_coverage": 0.46,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.4871,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.36,
"0.5": 0.54,
"0.75": 0.6133,
"0.9": 0.6333,
"1.0": 0.64
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.07,
"macro_f1": 0.0595,
"ece": 0.5381,
"brier": 1.3306,
"nll": 5.4549,
"mean_confidence": 0.6081,
"acc_at_50_coverage": 0.08,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8788,
"risk_at_coverage": {
"0.1": 0.8,
"0.25": 0.92,
"0.5": 0.92,
"0.75": 0.92,
"0.9": 0.9222,
"1.0": 0.93
}
},
"jev": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.914,
"ece": 0.0387,
"brier": 0.1328,
"nll": 1.0535,
"mean_confidence": 0.9156,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 0.009,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0333,
"1.0": 0.08
},
"latency_p50_ms": 224.4,
"latency_p95_ms": 373.2,
"cost_usd": 0.00263,
"input_tokens": 62684,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.36,
"kai~jev": 0.86,
"laya~jev": 0.33
}
},
"massive.hy": {
"kai": {
"n": 100,
"accuracy": 0.9,
"macro_f1": 0.9066,
"ece": 0.0642,
"brier": 0.1596,
"nll": 0.4587,
"mean_confidence": 0.9337,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0115,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0556,
"1.0": 0.1
}
},
"laya": {
"n": 100,
"accuracy": 0.25,
"macro_f1": 0.2364,
"ece": 0.5127,
"brier": 1.1723,
"nll": 4.1263,
"mean_confidence": 0.7627,
"acc_at_50_coverage": 0.3,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.7133,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.76,
"0.5": 0.7,
"0.75": 0.7067,
"0.9": 0.7222,
"1.0": 0.75
}
},
"laya:english": {
"n": 100,
"accuracy": 0.05,
"macro_f1": 0.0355,
"ece": 0.5709,
"brier": 1.4019,
"nll": 6.2584,
"mean_confidence": 0.6209,
"acc_at_50_coverage": 0.04,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.9629,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 0.96,
"0.5": 0.96,
"0.75": 0.9467,
"0.9": 0.9556,
"1.0": 0.95
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.25,
"macro_f1": 0.2364,
"ece": 0.5127,
"brier": 1.1723,
"nll": 4.1263,
"mean_confidence": 0.7627,
"acc_at_50_coverage": 0.3,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.7133,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.76,
"0.5": 0.7,
"0.75": 0.7067,
"0.9": 0.7222,
"1.0": 0.75
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.04,
"macro_f1": 0.0515,
"ece": 0.419,
"brier": 1.2273,
"nll": 4.4821,
"mean_confidence": 0.459,
"acc_at_50_coverage": 0.02,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.9777,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 0.96,
"0.5": 0.98,
"0.75": 0.9867,
"0.9": 0.9667,
"1.0": 0.96
}
},
"jev": {
"n": 100,
"accuracy": 0.84,
"macro_f1": 0.8592,
"ece": 0.0901,
"brier": 0.1861,
"nll": 0.6293,
"mean_confidence": 0.8733,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0227,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0333,
"0.9": 0.0778,
"1.0": 0.16
},
"latency_p50_ms": 225.6,
"latency_p95_ms": 320.7,
"cost_usd": 0.00268,
"input_tokens": 63845,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.27,
"kai~jev": 0.84,
"laya~jev": 0.25
}
},
"massive.id": {
"kai": {
"n": 100,
"accuracy": 0.96,
"macro_f1": 0.9549,
"ece": 0.0478,
"brier": 0.0745,
"nll": 0.1918,
"mean_confidence": 0.9646,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0064,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0111,
"1.0": 0.04
}
},
"laya": {
"n": 100,
"accuracy": 0.36,
"macro_f1": 0.3274,
"ece": 0.4856,
"brier": 1.1008,
"nll": 5.0724,
"mean_confidence": 0.8291,
"acc_at_50_coverage": 0.48,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.4555,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.28,
"0.5": 0.52,
"0.75": 0.6,
"0.9": 0.6333,
"1.0": 0.64
}
},
"laya:english": {
"n": 100,
"accuracy": 0.36,
"macro_f1": 0.3283,
"ece": 0.487,
"brier": 1.1037,
"nll": 5.0802,
"mean_confidence": 0.8304,
"acc_at_50_coverage": 0.48,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.4559,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.28,
"0.5": 0.52,
"0.75": 0.6,
"0.9": 0.6333,
"1.0": 0.64
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.51,
"macro_f1": 0.4973,
"ece": 0.3225,
"brier": 0.7524,
"nll": 2.2673,
"mean_confidence": 0.8281,
"acc_at_50_coverage": 0.76,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.239,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.12,
"0.5": 0.24,
"0.75": 0.3733,
"0.9": 0.4444,
"1.0": 0.49
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.35,
"macro_f1": 0.3399,
"ece": 0.3176,
"brier": 0.9377,
"nll": 3.0188,
"mean_confidence": 0.6676,
"acc_at_50_coverage": 0.46,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.4618,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.36,
"0.5": 0.54,
"0.75": 0.6,
"0.9": 0.6333,
"1.0": 0.65
}
},
"jev": {
"n": 100,
"accuracy": 0.93,
"macro_f1": 0.9372,
"ece": 0.0453,
"brier": 0.1195,
"nll": 0.7266,
"mean_confidence": 0.9333,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0092,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0444,
"1.0": 0.07
},
"latency_p50_ms": 232.2,
"latency_p95_ms": 327.9,
"cost_usd": 0.00261,
"input_tokens": 62203,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.36,
"kai~jev": 0.9,
"laya~jev": 0.39
}
},
"massive.is": {
"kai": {
"n": 100,
"accuracy": 0.86,
"macro_f1": 0.848,
"ece": 0.078,
"brier": 0.1946,
"nll": 0.4992,
"mean_confidence": 0.9217,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0161,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0778,
"1.0": 0.14
}
},
"laya": {
"n": 100,
"accuracy": 0.33,
"macro_f1": 0.3115,
"ece": 0.486,
"brier": 1.0847,
"nll": 4.8372,
"mean_confidence": 0.8091,
"acc_at_50_coverage": 0.4,
"unanswered": 0,
"zero_prob": 0.08,
"questions": 100,
"aurc": 0.5699,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.48,
"0.5": 0.6,
"0.75": 0.6133,
"0.9": 0.6333,
"1.0": 0.67
}
},
"laya:english": {
"n": 100,
"accuracy": 0.11,
"macro_f1": 0.0893,
"ece": 0.6641,
"brier": 1.4728,
"nll": 8.4791,
"mean_confidence": 0.7741,
"acc_at_50_coverage": 0.16,
"unanswered": 0,
"zero_prob": 0.16,
"questions": 100,
"aurc": 0.8328,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.84,
"0.5": 0.84,
"0.75": 0.8533,
"0.9": 0.8778,
"1.0": 0.89
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.35,
"macro_f1": 0.3355,
"ece": 0.4518,
"brier": 1.0317,
"nll": 4.0166,
"mean_confidence": 0.7949,
"acc_at_50_coverage": 0.46,
"unanswered": 0,
"zero_prob": 0.04,
"questions": 100,
"aurc": 0.4926,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.4,
"0.5": 0.54,
"0.75": 0.5867,
"0.9": 0.6222,
"1.0": 0.65
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.1,
"macro_f1": 0.0695,
"ece": 0.5395,
"brier": 1.3122,
"nll": 5.2892,
"mean_confidence": 0.6395,
"acc_at_50_coverage": 0.14,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8442,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.8,
"0.5": 0.86,
"0.75": 0.8933,
"0.9": 0.8889,
"1.0": 0.9
}
},
"jev": {
"n": 100,
"accuracy": 0.84,
"macro_f1": 0.8292,
"ece": 0.073,
"brier": 0.2212,
"nll": 0.9779,
"mean_confidence": 0.8445,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0248,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.04,
"0.9": 0.0889,
"1.0": 0.16
},
"latency_p50_ms": 230.6,
"latency_p95_ms": 364.7,
"cost_usd": 0.00263,
"input_tokens": 62574,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.35,
"kai~jev": 0.8,
"laya~jev": 0.38
}
},
"massive.it": {
"kai": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.9039,
"ece": 0.0604,
"brier": 0.1277,
"nll": 0.3608,
"mean_confidence": 0.9602,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0116,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.02,
"0.75": 0.0133,
"0.9": 0.0222,
"1.0": 0.08
}
},
"laya": {
"n": 100,
"accuracy": 0.47,
"macro_f1": 0.4507,
"ece": 0.38,
"brier": 0.8862,
"nll": 3.9536,
"mean_confidence": 0.8293,
"acc_at_50_coverage": 0.62,
"unanswered": 0,
"zero_prob": 0.07,
"questions": 100,
"aurc": 0.3937,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.36,
"0.5": 0.38,
"0.75": 0.4533,
"0.9": 0.4889,
"1.0": 0.53
}
},
"laya:english": {
"n": 100,
"accuracy": 0.34,
"macro_f1": 0.3101,
"ece": 0.5156,
"brier": 1.1115,
"nll": 5.5529,
"mean_confidence": 0.8534,
"acc_at_50_coverage": 0.52,
"unanswered": 0,
"zero_prob": 0.1,
"questions": 100,
"aurc": 0.4051,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.28,
"0.5": 0.48,
"0.75": 0.5733,
"0.9": 0.6333,
"1.0": 0.66
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.52,
"macro_f1": 0.5034,
"ece": 0.3428,
"brier": 0.7809,
"nll": 2.6981,
"mean_confidence": 0.819,
"acc_at_50_coverage": 0.68,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.3414,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.28,
"0.5": 0.32,
"0.75": 0.4133,
"0.9": 0.4444,
"1.0": 0.48
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.41,
"macro_f1": 0.3731,
"ece": 0.3423,
"brier": 0.8979,
"nll": 3.1843,
"mean_confidence": 0.7334,
"acc_at_50_coverage": 0.56,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.4054,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.32,
"0.5": 0.44,
"0.75": 0.52,
"0.9": 0.5889,
"1.0": 0.59
}
},
"jev": {
"n": 100,
"accuracy": 0.96,
"macro_f1": 0.9562,
"ece": 0.0502,
"brier": 0.0568,
"nll": 0.3569,
"mean_confidence": 0.9396,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0047,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0111,
"1.0": 0.04
},
"latency_p50_ms": 240.8,
"latency_p95_ms": 321.4,
"cost_usd": 0.00262,
"input_tokens": 62273,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.49,
"kai~jev": 0.9,
"laya~jev": 0.48
}
},
"massive.ja": {
"kai": {
"n": 100,
"accuracy": 0.97,
"macro_f1": 0.9673,
"ece": 0.0358,
"brier": 0.0675,
"nll": 0.2089,
"mean_confidence": 0.9664,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0022,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.03
}
},
"laya": {
"n": 100,
"accuracy": 0.64,
"macro_f1": 0.5896,
"ece": 0.2337,
"brier": 0.5645,
"nll": 1.7428,
"mean_confidence": 0.8737,
"acc_at_50_coverage": 0.84,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.1879,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.08,
"0.5": 0.16,
"0.75": 0.2533,
"0.9": 0.3222,
"1.0": 0.36
}
},
"laya:english": {
"n": 100,
"accuracy": 0.53,
"macro_f1": 0.4942,
"ece": 0.3657,
"brier": 0.832,
"nll": 3.7558,
"mean_confidence": 0.8957,
"acc_at_50_coverage": 0.76,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.2844,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.12,
"0.5": 0.24,
"0.75": 0.3733,
"0.9": 0.4444,
"1.0": 0.47
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.64,
"macro_f1": 0.5896,
"ece": 0.2337,
"brier": 0.5645,
"nll": 1.7428,
"mean_confidence": 0.8737,
"acc_at_50_coverage": 0.84,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.1879,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.08,
"0.5": 0.16,
"0.75": 0.2533,
"0.9": 0.3222,
"1.0": 0.36
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.56,
"macro_f1": 0.5347,
"ece": 0.2391,
"brier": 0.6878,
"nll": 2.1149,
"mean_confidence": 0.7824,
"acc_at_50_coverage": 0.78,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.2428,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.16,
"0.5": 0.22,
"0.75": 0.3467,
"0.9": 0.4,
"1.0": 0.44
}
},
"jev": {
"n": 100,
"accuracy": 0.97,
"macro_f1": 0.9545,
"ece": 0.0587,
"brier": 0.0602,
"nll": 0.3747,
"mean_confidence": 0.9431,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0035,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0095,
"0.9": 0.0111,
"1.0": 0.03
},
"latency_p50_ms": 243.4,
"latency_p95_ms": 351.8,
"cost_usd": 0.00263,
"input_tokens": 62666,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.62,
"kai~jev": 0.95,
"laya~jev": 0.64
}
},
"massive.jv": {
"kai": {
"n": 100,
"accuracy": 0.89,
"macro_f1": 0.8821,
"ece": 0.0868,
"brier": 0.1862,
"nll": 0.5787,
"mean_confidence": 0.9175,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0156,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0778,
"1.0": 0.11
}
},
"laya": {
"n": 100,
"accuracy": 0.16,
"macro_f1": 0.1541,
"ece": 0.6634,
"brier": 1.4425,
"nll": 7.5006,
"mean_confidence": 0.8234,
"acc_at_50_coverage": 0.16,
"unanswered": 0,
"zero_prob": 0.11,
"questions": 100,
"aurc": 0.7915,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.84,
"0.5": 0.84,
"0.75": 0.8133,
"0.9": 0.8333,
"1.0": 0.84
}
},
"laya:english": {
"n": 100,
"accuracy": 0.16,
"macro_f1": 0.1537,
"ece": 0.67,
"brier": 1.456,
"nll": 7.6072,
"mean_confidence": 0.83,
"acc_at_50_coverage": 0.16,
"unanswered": 0,
"zero_prob": 0.11,
"questions": 100,
"aurc": 0.7935,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.84,
"0.5": 0.84,
"0.75": 0.8133,
"0.9": 0.8333,
"1.0": 0.84
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.3,
"macro_f1": 0.2714,
"ece": 0.4842,
"brier": 1.1318,
"nll": 4.47,
"mean_confidence": 0.7676,
"acc_at_50_coverage": 0.4,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 0.6118,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.64,
"0.5": 0.6,
"0.75": 0.6667,
"0.9": 0.6889,
"1.0": 0.7
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.11,
"macro_f1": 0.0936,
"ece": 0.5344,
"brier": 1.2828,
"nll": 4.825,
"mean_confidence": 0.6444,
"acc_at_50_coverage": 0.16,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.792,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.76,
"0.5": 0.84,
"0.75": 0.8667,
"0.9": 0.8778,
"1.0": 0.89
}
},
"jev": {
"n": 100,
"accuracy": 0.75,
"macro_f1": 0.7285,
"ece": 0.0856,
"brier": 0.3814,
"nll": 1.8353,
"mean_confidence": 0.7758,
"acc_at_50_coverage": 0.9,
"unanswered": 0,
"zero_prob": 0.04,
"questions": 100,
"aurc": 0.0908,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.1,
"0.75": 0.1467,
"0.9": 0.2,
"1.0": 0.25
},
"latency_p50_ms": 247.3,
"latency_p95_ms": 334.0,
"cost_usd": 0.00262,
"input_tokens": 62277,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.17,
"kai~jev": 0.74,
"laya~jev": 0.2
}
},
"massive.ka": {
"kai": {
"n": 100,
"accuracy": 0.83,
"macro_f1": 0.8118,
"ece": 0.0756,
"brier": 0.2419,
"nll": 0.6848,
"mean_confidence": 0.8967,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0349,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.02,
"0.75": 0.04,
"0.9": 0.0889,
"1.0": 0.17
}
},
"laya": {
"n": 100,
"accuracy": 0.15,
"macro_f1": 0.1449,
"ece": 0.5422,
"brier": 1.2966,
"nll": 4.3721,
"mean_confidence": 0.6835,
"acc_at_50_coverage": 0.22,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8097,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.8,
"0.5": 0.78,
"0.75": 0.8267,
"0.9": 0.8444,
"1.0": 0.85
}
},
"laya:english": {
"n": 100,
"accuracy": 0.09,
"macro_f1": 0.0742,
"ece": 0.6102,
"brier": 1.4279,
"nll": 6.2363,
"mean_confidence": 0.7002,
"acc_at_50_coverage": 0.1,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.9134,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.92,
"0.5": 0.9,
"0.75": 0.9067,
"0.9": 0.9,
"1.0": 0.91
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.15,
"macro_f1": 0.1449,
"ece": 0.5422,
"brier": 1.2966,
"nll": 4.3721,
"mean_confidence": 0.6835,
"acc_at_50_coverage": 0.22,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8097,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.8,
"0.5": 0.78,
"0.75": 0.8267,
"0.9": 0.8444,
"1.0": 0.85
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.04,
"macro_f1": 0.0319,
"ece": 0.4568,
"brier": 1.2406,
"nll": 4.3894,
"mean_confidence": 0.4968,
"acc_at_50_coverage": 0.06,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.9532,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 0.92,
"0.5": 0.94,
"0.75": 0.9467,
"0.9": 0.9556,
"1.0": 0.96
}
},
"jev": {
"n": 100,
"accuracy": 0.88,
"macro_f1": 0.8594,
"ece": 0.1081,
"brier": 0.2376,
"nll": 0.9616,
"mean_confidence": 0.8629,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.039,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0933,
"0.9": 0.1111,
"1.0": 0.12
},
"latency_p50_ms": 225.1,
"latency_p95_ms": 364.0,
"cost_usd": 0.00267,
"input_tokens": 63634,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.15,
"kai~jev": 0.78,
"laya~jev": 0.15
}
},
"massive.km": {
"kai": {
"n": 100,
"accuracy": 0.85,
"macro_f1": 0.8438,
"ece": 0.0955,
"brier": 0.2378,
"nll": 0.5424,
"mean_confidence": 0.8864,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0315,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.1222,
"1.0": 0.15
}
},
"laya": {
"n": 100,
"accuracy": 0.2,
"macro_f1": 0.188,
"ece": 0.4454,
"brier": 1.1586,
"nll": 3.7259,
"mean_confidence": 0.6454,
"acc_at_50_coverage": 0.22,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.6855,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.64,
"0.5": 0.78,
"0.75": 0.7467,
"0.9": 0.7778,
"1.0": 0.8
}
},
"laya:english": {
"n": 100,
"accuracy": 0.0,
"macro_f1": 0.0,
"ece": 0.7051,
"brier": 1.5573,
"nll": 7.4182,
"mean_confidence": 0.7051,
"acc_at_50_coverage": 0.0,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 1.0,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 1.0,
"0.5": 1.0,
"0.75": 1.0,
"0.9": 1.0,
"1.0": 1.0
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.2,
"macro_f1": 0.188,
"ece": 0.4454,
"brier": 1.1586,
"nll": 3.7259,
"mean_confidence": 0.6454,
"acc_at_50_coverage": 0.22,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.6855,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.64,
"0.5": 0.78,
"0.75": 0.7467,
"0.9": 0.7778,
"1.0": 0.8
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.06,
"macro_f1": 0.0639,
"ece": 0.3552,
"brier": 1.1745,
"nll": 4.2697,
"mean_confidence": 0.4152,
"acc_at_50_coverage": 0.04,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.9687,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 1.0,
"0.5": 0.96,
"0.75": 0.9467,
"0.9": 0.9444,
"1.0": 0.94
}
},
"jev": {
"n": 100,
"accuracy": 0.81,
"macro_f1": 0.795,
"ece": 0.0731,
"brier": 0.2688,
"nll": 1.2603,
"mean_confidence": 0.8377,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 0.0484,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.01,
"0.75": 0.0933,
"0.9": 0.1444,
"1.0": 0.19
},
"latency_p50_ms": 229.6,
"latency_p95_ms": 325.2,
"cost_usd": 0.0027,
"input_tokens": 64279,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.17,
"kai~jev": 0.79,
"laya~jev": 0.16
}
},
"massive.kn": {
"kai": {
"n": 100,
"accuracy": 0.9,
"macro_f1": 0.8935,
"ece": 0.0775,
"brier": 0.1619,
"nll": 0.4362,
"mean_confidence": 0.9329,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0137,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0556,
"1.0": 0.1
}
},
"laya": {
"n": 100,
"accuracy": 0.3,
"macro_f1": 0.2465,
"ece": 0.388,
"brier": 1.028,
"nll": 3.3642,
"mean_confidence": 0.6578,
"acc_at_50_coverage": 0.36,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.6162,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.56,
"0.5": 0.64,
"0.75": 0.6533,
"0.9": 0.6889,
"1.0": 0.7
}
},
"laya:english": {
"n": 100,
"accuracy": 0.11,
"macro_f1": 0.0913,
"ece": 0.5903,
"brier": 1.4055,
"nll": 6.2851,
"mean_confidence": 0.6951,
"acc_at_50_coverage": 0.08,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.9276,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 0.92,
"0.5": 0.92,
"0.75": 0.9067,
"0.9": 0.9,
"1.0": 0.89
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.3,
"macro_f1": 0.2465,
"ece": 0.388,
"brier": 1.028,
"nll": 3.3642,
"mean_confidence": 0.6578,
"acc_at_50_coverage": 0.36,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.6162,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.56,
"0.5": 0.64,
"0.75": 0.6533,
"0.9": 0.6889,
"1.0": 0.7
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.12,
"macro_f1": 0.1016,
"ece": 0.3606,
"brier": 1.1718,
"nll": 4.1668,
"mean_confidence": 0.4806,
"acc_at_50_coverage": 0.12,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8859,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 0.8,
"0.5": 0.88,
"0.75": 0.88,
"0.9": 0.8778,
"1.0": 0.88
}
},
"jev": {
"n": 100,
"accuracy": 0.87,
"macro_f1": 0.8617,
"ece": 0.07,
"brier": 0.158,
"nll": 0.5547,
"mean_confidence": 0.8984,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0211,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.04,
"0.9": 0.0556,
"1.0": 0.13
},
"latency_p50_ms": 236.9,
"latency_p95_ms": 339.2,
"cost_usd": 0.00269,
"input_tokens": 64112,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.29,
"kai~jev": 0.82,
"laya~jev": 0.27
}
},
"massive.ko": {
"kai": {
"n": 100,
"accuracy": 0.91,
"macro_f1": 0.9111,
"ece": 0.0445,
"brier": 0.1251,
"nll": 0.2675,
"mean_confidence": 0.945,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0075,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0222,
"1.0": 0.09
}
},
"laya": {
"n": 100,
"accuracy": 0.47,
"macro_f1": 0.4428,
"ece": 0.3108,
"brier": 0.7771,
"nll": 2.34,
"mean_confidence": 0.7773,
"acc_at_50_coverage": 0.68,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.2997,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.16,
"0.5": 0.32,
"0.75": 0.4133,
"0.9": 0.4778,
"1.0": 0.53
}
},
"laya:english": {
"n": 100,
"accuracy": 0.11,
"macro_f1": 0.0943,
"ece": 0.6804,
"brier": 1.5102,
"nll": 7.6867,
"mean_confidence": 0.7904,
"acc_at_50_coverage": 0.14,
"unanswered": 0,
"zero_prob": 0.12,
"questions": 100,
"aurc": 0.8448,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.8,
"0.5": 0.86,
"0.75": 0.88,
"0.9": 0.8778,
"1.0": 0.89
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.47,
"macro_f1": 0.4428,
"ece": 0.3108,
"brier": 0.7771,
"nll": 2.34,
"mean_confidence": 0.7773,
"acc_at_50_coverage": 0.68,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.2997,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.16,
"0.5": 0.32,
"0.75": 0.4133,
"0.9": 0.4778,
"1.0": 0.53
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.06,
"macro_f1": 0.0519,
"ece": 0.5059,
"brier": 1.2604,
"nll": 4.6336,
"mean_confidence": 0.5659,
"acc_at_50_coverage": 0.12,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8776,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.84,
"0.5": 0.88,
"0.75": 0.92,
"0.9": 0.9333,
"1.0": 0.94
}
},
"jev": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9456,
"ece": 0.0423,
"brier": 0.0812,
"nll": 0.3968,
"mean_confidence": 0.9487,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.007,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0111,
"1.0": 0.06
},
"latency_p50_ms": 270.2,
"latency_p95_ms": 388.1,
"cost_usd": 0.00262,
"input_tokens": 62368,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.45,
"kai~jev": 0.91,
"laya~jev": 0.43
}
},
"massive.lv": {
"kai": {
"n": 100,
"accuracy": 0.89,
"macro_f1": 0.8861,
"ece": 0.0831,
"brier": 0.1648,
"nll": 0.4598,
"mean_confidence": 0.942,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0111,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0444,
"1.0": 0.11
}
},
"laya": {
"n": 100,
"accuracy": 0.3,
"macro_f1": 0.2899,
"ece": 0.486,
"brier": 1.1049,
"nll": 3.9548,
"mean_confidence": 0.786,
"acc_at_50_coverage": 0.4,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.5448,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.44,
"0.5": 0.6,
"0.75": 0.6533,
"0.9": 0.6667,
"1.0": 0.7
}
},
"laya:english": {
"n": 100,
"accuracy": 0.1,
"macro_f1": 0.0786,
"ece": 0.6659,
"brier": 1.4871,
"nll": 7.3307,
"mean_confidence": 0.7659,
"acc_at_50_coverage": 0.1,
"unanswered": 0,
"zero_prob": 0.08,
"questions": 100,
"aurc": 0.885,
"risk_at_coverage": {
"0.1": 0.8,
"0.25": 0.88,
"0.5": 0.9,
"0.75": 0.92,
"0.9": 0.9,
"1.0": 0.9
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.31,
"macro_f1": 0.3022,
"ece": 0.472,
"brier": 1.0767,
"nll": 3.7436,
"mean_confidence": 0.782,
"acc_at_50_coverage": 0.46,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.5248,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.44,
"0.5": 0.54,
"0.75": 0.64,
"0.9": 0.6556,
"1.0": 0.69
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.09,
"macro_f1": 0.0805,
"ece": 0.4966,
"brier": 1.2831,
"nll": 5.0617,
"mean_confidence": 0.5866,
"acc_at_50_coverage": 0.12,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8759,
"risk_at_coverage": {
"0.1": 0.8,
"0.25": 0.92,
"0.5": 0.88,
"0.75": 0.92,
"0.9": 0.9111,
"1.0": 0.91
}
},
"jev": {
"n": 100,
"accuracy": 0.85,
"macro_f1": 0.8447,
"ece": 0.0748,
"brier": 0.2047,
"nll": 0.935,
"mean_confidence": 0.86,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0253,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.1,
"1.0": 0.15
},
"latency_p50_ms": 246.6,
"latency_p95_ms": 330.0,
"cost_usd": 0.00263,
"input_tokens": 62691,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.32,
"kai~jev": 0.82,
"laya~jev": 0.3
}
},
"massive.ml": {
"kai": {
"n": 100,
"accuracy": 0.88,
"macro_f1": 0.8692,
"ece": 0.094,
"brier": 0.217,
"nll": 0.6435,
"mean_confidence": 0.9206,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0258,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0889,
"1.0": 0.12
}
},
"laya": {
"n": 100,
"accuracy": 0.28,
"macro_f1": 0.2712,
"ece": 0.4433,
"brier": 1.0956,
"nll": 3.5857,
"mean_confidence": 0.7072,
"acc_at_50_coverage": 0.32,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.6375,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.6,
"0.5": 0.68,
"0.75": 0.6667,
"0.9": 0.7111,
"1.0": 0.72
}
},
"laya:english": {
"n": 100,
"accuracy": 0.07,
"macro_f1": 0.0503,
"ece": 0.601,
"brier": 1.3834,
"nll": 6.2205,
"mean_confidence": 0.671,
"acc_at_50_coverage": 0.1,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.8316,
"risk_at_coverage": {
"0.1": 0.7,
"0.25": 0.84,
"0.5": 0.9,
"0.75": 0.9067,
"0.9": 0.9222,
"1.0": 0.93
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.28,
"macro_f1": 0.2712,
"ece": 0.4433,
"brier": 1.0956,
"nll": 3.5857,
"mean_confidence": 0.7072,
"acc_at_50_coverage": 0.32,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.6375,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.6,
"0.5": 0.68,
"0.75": 0.6667,
"0.9": 0.7111,
"1.0": 0.72
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.1,
"macro_f1": 0.1013,
"ece": 0.3686,
"brier": 1.1473,
"nll": 4.1697,
"mean_confidence": 0.4686,
"acc_at_50_coverage": 0.1,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8208,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.84,
"0.5": 0.9,
"0.75": 0.9067,
"0.9": 0.8889,
"1.0": 0.9
}
},
"jev": {
"n": 100,
"accuracy": 0.91,
"macro_f1": 0.9029,
"ece": 0.0495,
"brier": 0.1368,
"nll": 0.5517,
"mean_confidence": 0.9051,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0111,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0389,
"1.0": 0.09
},
"latency_p50_ms": 258.4,
"latency_p95_ms": 377.3,
"cost_usd": 0.00271,
"input_tokens": 64495,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.26,
"kai~jev": 0.87,
"laya~jev": 0.27
}
},
"massive.mn": {
"kai": {
"n": 100,
"accuracy": 0.89,
"macro_f1": 0.8792,
"ece": 0.0943,
"brier": 0.1972,
"nll": 0.5781,
"mean_confidence": 0.9431,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0191,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0778,
"1.0": 0.11
}
},
"laya": {
"n": 100,
"accuracy": 0.16,
"macro_f1": 0.1422,
"ece": 0.577,
"brier": 1.3206,
"nll": 5.0728,
"mean_confidence": 0.737,
"acc_at_50_coverage": 0.22,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 0.7486,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.72,
"0.5": 0.78,
"0.75": 0.7867,
"0.9": 0.8222,
"1.0": 0.84
}
},
"laya:english": {
"n": 100,
"accuracy": 0.13,
"macro_f1": 0.1213,
"ece": 0.6704,
"brier": 1.4925,
"nll": 7.9451,
"mean_confidence": 0.7977,
"acc_at_50_coverage": 0.14,
"unanswered": 0,
"zero_prob": 0.1,
"questions": 100,
"aurc": 0.887,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 0.84,
"0.5": 0.86,
"0.75": 0.8933,
"0.9": 0.8889,
"1.0": 0.87
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.16,
"macro_f1": 0.1422,
"ece": 0.577,
"brier": 1.3206,
"nll": 5.0728,
"mean_confidence": 0.737,
"acc_at_50_coverage": 0.22,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 0.7486,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.72,
"0.5": 0.78,
"0.75": 0.7867,
"0.9": 0.8222,
"1.0": 0.84
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.07,
"macro_f1": 0.0595,
"ece": 0.569,
"brier": 1.3665,
"nll": 5.2813,
"mean_confidence": 0.639,
"acc_at_50_coverage": 0.08,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8957,
"risk_at_coverage": {
"0.1": 0.8,
"0.25": 0.88,
"0.5": 0.92,
"0.75": 0.9333,
"0.9": 0.9333,
"1.0": 0.93
}
},
"jev": {
"n": 100,
"accuracy": 0.84,
"macro_f1": 0.8334,
"ece": 0.0825,
"brier": 0.2063,
"nll": 0.6907,
"mean_confidence": 0.8655,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0252,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0889,
"1.0": 0.16
},
"latency_p50_ms": 236.4,
"latency_p95_ms": 341.8,
"cost_usd": 0.00269,
"input_tokens": 64102,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.18,
"kai~jev": 0.83,
"laya~jev": 0.17
}
},
"massive.ms": {
"kai": {
"n": 100,
"accuracy": 0.93,
"macro_f1": 0.9106,
"ece": 0.0467,
"brier": 0.1141,
"nll": 0.3493,
"mean_confidence": 0.9635,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0112,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.02,
"0.75": 0.0133,
"0.9": 0.0222,
"1.0": 0.07
}
},
"laya": {
"n": 100,
"accuracy": 0.27,
"macro_f1": 0.2372,
"ece": 0.5064,
"brier": 1.1586,
"nll": 5.3374,
"mean_confidence": 0.7764,
"acc_at_50_coverage": 0.4,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.546,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.36,
"0.5": 0.6,
"0.75": 0.6933,
"0.9": 0.7111,
"1.0": 0.73
}
},
"laya:english": {
"n": 100,
"accuracy": 0.27,
"macro_f1": 0.2372,
"ece": 0.5064,
"brier": 1.1586,
"nll": 5.3374,
"mean_confidence": 0.7764,
"acc_at_50_coverage": 0.4,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.546,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.36,
"0.5": 0.6,
"0.75": 0.6933,
"0.9": 0.7111,
"1.0": 0.73
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.41,
"macro_f1": 0.3787,
"ece": 0.4364,
"brier": 0.9581,
"nll": 3.3356,
"mean_confidence": 0.8269,
"acc_at_50_coverage": 0.58,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.4225,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.4,
"0.5": 0.42,
"0.75": 0.5467,
"0.9": 0.5556,
"1.0": 0.59
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.24,
"macro_f1": 0.1995,
"ece": 0.3277,
"brier": 1.0249,
"nll": 3.3724,
"mean_confidence": 0.5672,
"acc_at_50_coverage": 0.3,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.6168,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.6,
"0.5": 0.7,
"0.75": 0.72,
"0.9": 0.7444,
"1.0": 0.76
}
},
"jev": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.9192,
"ece": 0.0684,
"brier": 0.1374,
"nll": 1.0285,
"mean_confidence": 0.92,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 0.0143,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0333,
"0.9": 0.0333,
"1.0": 0.08
},
"latency_p50_ms": 234.3,
"latency_p95_ms": 427.7,
"cost_usd": 0.00262,
"input_tokens": 62274,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.29,
"kai~jev": 0.89,
"laya~jev": 0.27
}
},
"massive.my": {
"kai": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.8948,
"ece": 0.0656,
"brier": 0.1119,
"nll": 0.316,
"mean_confidence": 0.9351,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0117,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0222,
"1.0": 0.08
}
},
"laya": {
"n": 100,
"accuracy": 0.16,
"macro_f1": 0.1424,
"ece": 0.4766,
"brier": 1.2443,
"nll": 3.989,
"mean_confidence": 0.6219,
"acc_at_50_coverage": 0.1,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8866,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.92,
"0.5": 0.9,
"0.75": 0.84,
"0.9": 0.8444,
"1.0": 0.84
}
},
"laya:english": {
"n": 100,
"accuracy": 0.06,
"macro_f1": 0.0455,
"ece": 0.5797,
"brier": 1.3653,
"nll": 5.8399,
"mean_confidence": 0.6397,
"acc_at_50_coverage": 0.08,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.9128,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.88,
"0.5": 0.92,
"0.75": 0.92,
"0.9": 0.9333,
"1.0": 0.94
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.16,
"macro_f1": 0.1424,
"ece": 0.4766,
"brier": 1.2443,
"nll": 3.989,
"mean_confidence": 0.6219,
"acc_at_50_coverage": 0.1,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8866,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.92,
"0.5": 0.9,
"0.75": 0.84,
"0.9": 0.8444,
"1.0": 0.84
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.08,
"macro_f1": 0.0699,
"ece": 0.3454,
"brier": 1.1473,
"nll": 4.0301,
"mean_confidence": 0.4254,
"acc_at_50_coverage": 0.1,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.9293,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 1.0,
"0.5": 0.9,
"0.75": 0.9067,
"0.9": 0.9111,
"1.0": 0.92
}
},
"jev": {
"n": 100,
"accuracy": 0.89,
"macro_f1": 0.8853,
"ece": 0.0572,
"brier": 0.1661,
"nll": 0.5575,
"mean_confidence": 0.8781,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0182,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0778,
"1.0": 0.11
},
"latency_p50_ms": 262.8,
"latency_p95_ms": 390.2,
"cost_usd": 0.00271,
"input_tokens": 64450,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.17,
"kai~jev": 0.84,
"laya~jev": 0.16
}
},
"massive.nb": {
"kai": {
"n": 100,
"accuracy": 0.95,
"macro_f1": 0.9429,
"ece": 0.04,
"brier": 0.0839,
"nll": 0.2553,
"mean_confidence": 0.9761,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0045,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0111,
"1.0": 0.05
}
},
"laya": {
"n": 100,
"accuracy": 0.5,
"macro_f1": 0.4519,
"ece": 0.3679,
"brier": 0.8543,
"nll": 3.8176,
"mean_confidence": 0.8679,
"acc_at_50_coverage": 0.68,
"unanswered": 0,
"zero_prob": 0.04,
"questions": 100,
"aurc": 0.2972,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.2,
"0.5": 0.32,
"0.75": 0.44,
"0.9": 0.4667,
"1.0": 0.5
}
},
"laya:english": {
"n": 100,
"accuracy": 0.33,
"macro_f1": 0.3025,
"ece": 0.5179,
"brier": 1.1373,
"nll": 5.8232,
"mean_confidence": 0.8479,
"acc_at_50_coverage": 0.52,
"unanswered": 0,
"zero_prob": 0.09,
"questions": 100,
"aurc": 0.4227,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.32,
"0.5": 0.48,
"0.75": 0.5733,
"0.9": 0.6444,
"1.0": 0.67
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.56,
"macro_f1": 0.5174,
"ece": 0.2937,
"brier": 0.7025,
"nll": 2.4173,
"mean_confidence": 0.8537,
"acc_at_50_coverage": 0.74,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.2423,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.2,
"0.5": 0.26,
"0.75": 0.3467,
"0.9": 0.3889,
"1.0": 0.44
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.36,
"macro_f1": 0.317,
"ece": 0.3212,
"brier": 0.8917,
"nll": 3.295,
"mean_confidence": 0.6812,
"acc_at_50_coverage": 0.56,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.3869,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.28,
"0.5": 0.44,
"0.75": 0.5333,
"0.9": 0.6,
"1.0": 0.64
}
},
"jev": {
"n": 100,
"accuracy": 0.86,
"macro_f1": 0.8364,
"ece": 0.0713,
"brier": 0.1846,
"nll": 0.6197,
"mean_confidence": 0.8865,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0223,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0778,
"1.0": 0.14
},
"latency_p50_ms": 241.2,
"latency_p95_ms": 370.0,
"cost_usd": 0.00262,
"input_tokens": 62273,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.5,
"kai~jev": 0.84,
"laya~jev": 0.5
}
},
"massive.nl": {
"kai": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.9057,
"ece": 0.0537,
"brier": 0.1222,
"nll": 0.2807,
"mean_confidence": 0.9678,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.006,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0222,
"1.0": 0.08
}
},
"laya": {
"n": 100,
"accuracy": 0.4,
"macro_f1": 0.3967,
"ece": 0.4792,
"brier": 1.0453,
"nll": 5.621,
"mean_confidence": 0.8792,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.16,
"questions": 100,
"aurc": 0.445,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.4,
"0.5": 0.4,
"0.75": 0.5333,
"0.9": 0.5889,
"1.0": 0.6
}
},
"laya:english": {
"n": 100,
"accuracy": 0.39,
"macro_f1": 0.3968,
"ece": 0.5025,
"brier": 1.0787,
"nll": 6.068,
"mean_confidence": 0.8831,
"acc_at_50_coverage": 0.56,
"unanswered": 0,
"zero_prob": 0.18,
"questions": 100,
"aurc": 0.4603,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.4,
"0.5": 0.44,
"0.75": 0.56,
"0.9": 0.6111,
"1.0": 0.61
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.47,
"macro_f1": 0.4069,
"ece": 0.397,
"brier": 0.8739,
"nll": 2.9847,
"mean_confidence": 0.8557,
"acc_at_50_coverage": 0.68,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 0.3565,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.24,
"0.5": 0.32,
"0.75": 0.4667,
"0.9": 0.5111,
"1.0": 0.53
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.37,
"macro_f1": 0.3444,
"ece": 0.3445,
"brier": 0.924,
"nll": 3.2633,
"mean_confidence": 0.7145,
"acc_at_50_coverage": 0.48,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.4568,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.36,
"0.5": 0.52,
"0.75": 0.5733,
"0.9": 0.5889,
"1.0": 0.63
}
},
"jev": {
"n": 100,
"accuracy": 0.9,
"macro_f1": 0.8966,
"ece": 0.0551,
"brier": 0.1544,
"nll": 0.827,
"mean_confidence": 0.9165,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0142,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0444,
"1.0": 0.1
},
"latency_p50_ms": 238.3,
"latency_p95_ms": 375.1,
"cost_usd": 0.00262,
"input_tokens": 62294,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.38,
"kai~jev": 0.9,
"laya~jev": 0.37
}
},
"massive.pl": {
"kai": {
"n": 100,
"accuracy": 0.93,
"macro_f1": 0.9226,
"ece": 0.0483,
"brier": 0.1089,
"nll": 0.266,
"mean_confidence": 0.9603,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0051,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0111,
"1.0": 0.07
}
},
"laya": {
"n": 100,
"accuracy": 0.47,
"macro_f1": 0.4548,
"ece": 0.3737,
"brier": 0.8546,
"nll": 2.9006,
"mean_confidence": 0.8293,
"acc_at_50_coverage": 0.74,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.2628,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.04,
"0.5": 0.26,
"0.75": 0.44,
"0.9": 0.5111,
"1.0": 0.53
}
},
"laya:english": {
"n": 100,
"accuracy": 0.24,
"macro_f1": 0.2033,
"ece": 0.5695,
"brier": 1.2812,
"nll": 6.3695,
"mean_confidence": 0.7951,
"acc_at_50_coverage": 0.36,
"unanswered": 0,
"zero_prob": 0.08,
"questions": 100,
"aurc": 0.5929,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.52,
"0.5": 0.64,
"0.75": 0.72,
"0.9": 0.7556,
"1.0": 0.76
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.5,
"macro_f1": 0.4781,
"ece": 0.3529,
"brier": 0.7994,
"nll": 2.5359,
"mean_confidence": 0.8407,
"acc_at_50_coverage": 0.76,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.2589,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.04,
"0.5": 0.24,
"0.75": 0.4,
"0.9": 0.4778,
"1.0": 0.5
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.25,
"macro_f1": 0.1982,
"ece": 0.3442,
"brier": 1.0012,
"nll": 3.6854,
"mean_confidence": 0.5933,
"acc_at_50_coverage": 0.42,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.5573,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.44,
"0.5": 0.58,
"0.75": 0.7067,
"0.9": 0.7333,
"1.0": 0.75
}
},
"jev": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.9288,
"ece": 0.0358,
"brier": 0.1231,
"nll": 0.4784,
"mean_confidence": 0.9243,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0105,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0444,
"1.0": 0.08
},
"latency_p50_ms": 246.3,
"latency_p95_ms": 346.8,
"cost_usd": 0.00262,
"input_tokens": 62499,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.48,
"kai~jev": 0.9,
"laya~jev": 0.43
}
},
"massive.pt": {
"kai": {
"n": 100,
"accuracy": 0.93,
"macro_f1": 0.9191,
"ece": 0.0292,
"brier": 0.1041,
"nll": 0.3583,
"mean_confidence": 0.9497,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0065,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0111,
"1.0": 0.07
}
},
"laya": {
"n": 100,
"accuracy": 0.44,
"macro_f1": 0.4162,
"ece": 0.3956,
"brier": 0.9349,
"nll": 4.2072,
"mean_confidence": 0.8337,
"acc_at_50_coverage": 0.62,
"unanswered": 0,
"zero_prob": 0.08,
"questions": 100,
"aurc": 0.385,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.32,
"0.5": 0.38,
"0.75": 0.4933,
"0.9": 0.5333,
"1.0": 0.56
}
},
"laya:english": {
"n": 100,
"accuracy": 0.47,
"macro_f1": 0.4213,
"ece": 0.4317,
"brier": 0.9253,
"nll": 5.3812,
"mean_confidence": 0.8894,
"acc_at_50_coverage": 0.66,
"unanswered": 0,
"zero_prob": 0.15,
"questions": 100,
"aurc": 0.3301,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.24,
"0.5": 0.34,
"0.75": 0.4133,
"0.9": 0.5111,
"1.0": 0.53
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.5,
"macro_f1": 0.4598,
"ece": 0.3405,
"brier": 0.7884,
"nll": 2.5341,
"mean_confidence": 0.8136,
"acc_at_50_coverage": 0.72,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.2881,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.24,
"0.5": 0.28,
"0.75": 0.4267,
"0.9": 0.4667,
"1.0": 0.5
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.5,
"macro_f1": 0.4758,
"ece": 0.2695,
"brier": 0.7686,
"nll": 2.6742,
"mean_confidence": 0.7568,
"acc_at_50_coverage": 0.64,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.3309,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.32,
"0.5": 0.36,
"0.75": 0.44,
"0.9": 0.4556,
"1.0": 0.5
}
},
"jev": {
"n": 100,
"accuracy": 0.9,
"macro_f1": 0.9031,
"ece": 0.0985,
"brier": 0.159,
"nll": 1.0782,
"mean_confidence": 0.9111,
"acc_at_50_coverage": 0.96,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 0.0447,
"risk_at_coverage": {
"0.1": 0.0385,
"0.25": 0.0385,
"0.5": 0.0385,
"0.75": 0.0533,
"0.9": 0.0444,
"1.0": 0.1
},
"latency_p50_ms": 248.8,
"latency_p95_ms": 329.2,
"cost_usd": 0.00261,
"input_tokens": 62236,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.42,
"kai~jev": 0.89,
"laya~jev": 0.45
}
},
"massive.ro": {
"kai": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.9084,
"ece": 0.0528,
"brier": 0.0959,
"nll": 0.2437,
"mean_confidence": 0.9464,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0052,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0111,
"1.0": 0.08
}
},
"laya": {
"n": 100,
"accuracy": 0.37,
"macro_f1": 0.3513,
"ece": 0.473,
"brier": 1.0551,
"nll": 4.9228,
"mean_confidence": 0.843,
"acc_at_50_coverage": 0.52,
"unanswered": 0,
"zero_prob": 0.07,
"questions": 100,
"aurc": 0.4941,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.44,
"0.5": 0.48,
"0.75": 0.56,
"0.9": 0.6,
"1.0": 0.63
}
},
"laya:english": {
"n": 100,
"accuracy": 0.33,
"macro_f1": 0.2976,
"ece": 0.5499,
"brier": 1.144,
"nll": 5.9633,
"mean_confidence": 0.8668,
"acc_at_50_coverage": 0.48,
"unanswered": 0,
"zero_prob": 0.11,
"questions": 100,
"aurc": 0.4988,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.52,
"0.5": 0.52,
"0.75": 0.6,
"0.9": 0.6556,
"1.0": 0.67
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.37,
"macro_f1": 0.3426,
"ece": 0.3974,
"brier": 0.96,
"nll": 3.0397,
"mean_confidence": 0.7674,
"acc_at_50_coverage": 0.56,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.3993,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.24,
"0.5": 0.44,
"0.75": 0.56,
"0.9": 0.6,
"1.0": 0.63
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.31,
"macro_f1": 0.2675,
"ece": 0.399,
"brier": 1.0408,
"nll": 3.6373,
"mean_confidence": 0.6696,
"acc_at_50_coverage": 0.38,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.5802,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.52,
"0.5": 0.62,
"0.75": 0.6933,
"0.9": 0.6778,
"1.0": 0.69
}
},
"jev": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9302,
"ece": 0.0822,
"brier": 0.102,
"nll": 0.7529,
"mean_confidence": 0.9085,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0071,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0111,
"1.0": 0.06
},
"latency_p50_ms": 271.2,
"latency_p95_ms": 378.9,
"cost_usd": 0.00262,
"input_tokens": 62352,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.38,
"kai~jev": 0.91,
"laya~jev": 0.37
}
},
"massive.ru": {
"kai": {
"n": 100,
"accuracy": 0.97,
"macro_f1": 0.9656,
"ece": 0.016,
"brier": 0.0438,
"nll": 0.0994,
"mean_confidence": 0.9692,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0008,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.03
}
},
"laya": {
"n": 100,
"accuracy": 0.57,
"macro_f1": 0.5342,
"ece": 0.2959,
"brier": 0.7319,
"nll": 2.5767,
"mean_confidence": 0.8594,
"acc_at_50_coverage": 0.7,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.3078,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.24,
"0.5": 0.3,
"0.75": 0.3467,
"0.9": 0.3889,
"1.0": 0.43
}
},
"laya:english": {
"n": 100,
"accuracy": 0.31,
"macro_f1": 0.3084,
"ece": 0.5946,
"brier": 1.228,
"nll": 6.4115,
"mean_confidence": 0.9046,
"acc_at_50_coverage": 0.42,
"unanswered": 0,
"zero_prob": 0.15,
"questions": 100,
"aurc": 0.5355,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.44,
"0.5": 0.58,
"0.75": 0.6267,
"0.9": 0.6556,
"1.0": 0.69
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.57,
"macro_f1": 0.5342,
"ece": 0.2959,
"brier": 0.7319,
"nll": 2.5767,
"mean_confidence": 0.8594,
"acc_at_50_coverage": 0.7,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.3078,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.24,
"0.5": 0.3,
"0.75": 0.3467,
"0.9": 0.3889,
"1.0": 0.43
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.39,
"macro_f1": 0.3815,
"ece": 0.2835,
"brier": 0.9005,
"nll": 3.0722,
"mean_confidence": 0.6626,
"acc_at_50_coverage": 0.52,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.4751,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.4,
"0.5": 0.48,
"0.75": 0.52,
"0.9": 0.5778,
"1.0": 0.61
}
},
"jev": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9408,
"ece": 0.0325,
"brier": 0.0781,
"nll": 0.4076,
"mean_confidence": 0.9461,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0052,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0111,
"0.9": 0.0111,
"1.0": 0.06
},
"latency_p50_ms": 286.5,
"latency_p95_ms": 1282.4,
"cost_usd": 0.00265,
"input_tokens": 63197,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.56,
"kai~jev": 0.92,
"laya~jev": 0.54
}
},
"massive.sl": {
"kai": {
"n": 100,
"accuracy": 0.9,
"macro_f1": 0.8984,
"ece": 0.0438,
"brier": 0.1298,
"nll": 0.3176,
"mean_confidence": 0.9422,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0079,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0222,
"1.0": 0.1
}
},
"laya": {
"n": 100,
"accuracy": 0.33,
"macro_f1": 0.2924,
"ece": 0.4871,
"brier": 1.1102,
"nll": 4.9705,
"mean_confidence": 0.8171,
"acc_at_50_coverage": 0.4,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.5472,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.52,
"0.5": 0.6,
"0.75": 0.64,
"0.9": 0.6444,
"1.0": 0.67
}
},
"laya:english": {
"n": 100,
"accuracy": 0.2,
"macro_f1": 0.1679,
"ece": 0.5994,
"brier": 1.3626,
"nll": 7.5628,
"mean_confidence": 0.7994,
"acc_at_50_coverage": 0.26,
"unanswered": 0,
"zero_prob": 0.12,
"questions": 100,
"aurc": 0.6777,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.68,
"0.5": 0.74,
"0.75": 0.7867,
"0.9": 0.7889,
"1.0": 0.8
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.37,
"macro_f1": 0.348,
"ece": 0.4338,
"brier": 1.0103,
"nll": 3.2354,
"mean_confidence": 0.7952,
"acc_at_50_coverage": 0.48,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.4403,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.36,
"0.5": 0.52,
"0.75": 0.5867,
"0.9": 0.6,
"1.0": 0.63
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.16,
"macro_f1": 0.1254,
"ece": 0.3987,
"brier": 1.1234,
"nll": 4.4217,
"mean_confidence": 0.5587,
"acc_at_50_coverage": 0.22,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.6819,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.64,
"0.5": 0.78,
"0.75": 0.8,
"0.9": 0.8222,
"1.0": 0.84
}
},
"jev": {
"n": 100,
"accuracy": 0.89,
"macro_f1": 0.8893,
"ece": 0.0579,
"brier": 0.16,
"nll": 0.5963,
"mean_confidence": 0.9104,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0148,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0444,
"1.0": 0.11
},
"latency_p50_ms": 261.3,
"latency_p95_ms": 329.1,
"cost_usd": 0.00262,
"input_tokens": 62385,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.33,
"kai~jev": 0.9,
"laya~jev": 0.34
}
},
"massive.sq": {
"kai": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9293,
"ece": 0.0503,
"brier": 0.1077,
"nll": 0.3327,
"mean_confidence": 0.9331,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0046,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0111,
"1.0": 0.06
}
},
"laya": {
"n": 100,
"accuracy": 0.29,
"macro_f1": 0.2575,
"ece": 0.4996,
"brier": 1.1614,
"nll": 4.0241,
"mean_confidence": 0.7822,
"acc_at_50_coverage": 0.38,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.5334,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.44,
"0.5": 0.62,
"0.75": 0.6933,
"0.9": 0.7,
"1.0": 0.71
}
},
"laya:english": {
"n": 100,
"accuracy": 0.21,
"macro_f1": 0.1667,
"ece": 0.6031,
"brier": 1.3494,
"nll": 7.3943,
"mean_confidence": 0.8131,
"acc_at_50_coverage": 0.28,
"unanswered": 0,
"zero_prob": 0.16,
"questions": 100,
"aurc": 0.684,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.68,
"0.5": 0.72,
"0.75": 0.7733,
"0.9": 0.7889,
"1.0": 0.79
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.3,
"macro_f1": 0.2679,
"ece": 0.4869,
"brier": 1.1324,
"nll": 3.6543,
"mean_confidence": 0.7795,
"acc_at_50_coverage": 0.4,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.5115,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.4,
"0.5": 0.6,
"0.75": 0.6667,
"0.9": 0.6889,
"1.0": 0.7
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.19,
"macro_f1": 0.1637,
"ece": 0.4217,
"brier": 1.1237,
"nll": 4.275,
"mean_confidence": 0.6117,
"acc_at_50_coverage": 0.3,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.6277,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.52,
"0.5": 0.7,
"0.75": 0.76,
"0.9": 0.8,
"1.0": 0.81
}
},
"jev": {
"n": 100,
"accuracy": 0.76,
"macro_f1": 0.7662,
"ece": 0.094,
"brier": 0.3067,
"nll": 1.1755,
"mean_confidence": 0.8186,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0531,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.02,
"0.75": 0.0933,
"0.9": 0.1778,
"1.0": 0.24
},
"latency_p50_ms": 255.7,
"latency_p95_ms": 406.9,
"cost_usd": 0.00263,
"input_tokens": 62527,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.3,
"kai~jev": 0.77,
"laya~jev": 0.29
}
},
"massive.sv": {
"kai": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9406,
"ece": 0.0479,
"brier": 0.0992,
"nll": 0.2203,
"mean_confidence": 0.9633,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0039,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0111,
"1.0": 0.06
}
},
"laya": {
"n": 100,
"accuracy": 0.48,
"macro_f1": 0.4253,
"ece": 0.3699,
"brier": 0.8409,
"nll": 3.286,
"mean_confidence": 0.8481,
"acc_at_50_coverage": 0.7,
"unanswered": 0,
"zero_prob": 0.04,
"questions": 100,
"aurc": 0.3547,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.28,
"0.5": 0.3,
"0.75": 0.44,
"0.9": 0.4778,
"1.0": 0.52
}
},
"laya:english": {
"n": 100,
"accuracy": 0.38,
"macro_f1": 0.3809,
"ece": 0.4874,
"brier": 1.0817,
"nll": 5.4997,
"mean_confidence": 0.8501,
"acc_at_50_coverage": 0.5,
"unanswered": 0,
"zero_prob": 0.09,
"questions": 100,
"aurc": 0.445,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.32,
"0.5": 0.5,
"0.75": 0.6,
"0.9": 0.6111,
"1.0": 0.62
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.49,
"macro_f1": 0.4496,
"ece": 0.341,
"brier": 0.7772,
"nll": 2.587,
"mean_confidence": 0.8215,
"acc_at_50_coverage": 0.74,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.3089,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.2,
"0.5": 0.26,
"0.75": 0.3867,
"0.9": 0.4778,
"1.0": 0.51
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.24,
"macro_f1": 0.212,
"ece": 0.4338,
"brier": 1.0703,
"nll": 3.6524,
"mean_confidence": 0.6687,
"acc_at_50_coverage": 0.38,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.5683,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.52,
"0.5": 0.62,
"0.75": 0.7333,
"0.9": 0.7667,
"1.0": 0.76
}
},
"jev": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9311,
"ece": 0.0737,
"brier": 0.1188,
"nll": 0.5312,
"mean_confidence": 0.9187,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0103,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0222,
"1.0": 0.06
},
"latency_p50_ms": 242.8,
"latency_p95_ms": 334.8,
"cost_usd": 0.00262,
"input_tokens": 62329,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.49,
"kai~jev": 0.9,
"laya~jev": 0.46
}
},
"massive.sw": {
"kai": {
"n": 100,
"accuracy": 0.84,
"macro_f1": 0.8444,
"ece": 0.0829,
"brier": 0.2269,
"nll": 0.7429,
"mean_confidence": 0.9132,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0195,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0889,
"1.0": 0.16
}
},
"laya": {
"n": 100,
"accuracy": 0.13,
"macro_f1": 0.1113,
"ece": 0.6279,
"brier": 1.4219,
"nll": 6.7622,
"mean_confidence": 0.7471,
"acc_at_50_coverage": 0.16,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.8262,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.76,
"0.5": 0.84,
"0.75": 0.8533,
"0.9": 0.8667,
"1.0": 0.87
}
},
"laya:english": {
"n": 100,
"accuracy": 0.13,
"macro_f1": 0.1149,
"ece": 0.6318,
"brier": 1.4341,
"nll": 6.9309,
"mean_confidence": 0.7511,
"acc_at_50_coverage": 0.14,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.842,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.8,
"0.5": 0.86,
"0.75": 0.8533,
"0.9": 0.8778,
"1.0": 0.87
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.23,
"macro_f1": 0.2259,
"ece": 0.4922,
"brier": 1.1812,
"nll": 3.9796,
"mean_confidence": 0.7118,
"acc_at_50_coverage": 0.3,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.5978,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.52,
"0.5": 0.7,
"0.75": 0.7333,
"0.9": 0.7556,
"1.0": 0.77
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.1,
"macro_f1": 0.0772,
"ece": 0.4629,
"brier": 1.2274,
"nll": 4.7365,
"mean_confidence": 0.5629,
"acc_at_50_coverage": 0.16,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8354,
"risk_at_coverage": {
"0.1": 0.8,
"0.25": 0.84,
"0.5": 0.84,
"0.75": 0.8667,
"0.9": 0.8889,
"1.0": 0.9
}
},
"jev": {
"n": 100,
"accuracy": 0.7,
"macro_f1": 0.7043,
"ece": 0.1124,
"brier": 0.3897,
"nll": 2.6964,
"mean_confidence": 0.732,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.07,
"questions": 100,
"aurc": 0.0724,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.01,
"0.75": 0.1467,
"0.9": 0.2333,
"1.0": 0.3
},
"latency_p50_ms": 248.4,
"latency_p95_ms": 359.7,
"cost_usd": 0.00262,
"input_tokens": 62369,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.14,
"kai~jev": 0.72,
"laya~jev": 0.13
}
},
"massive.ta": {
"kai": {
"n": 100,
"accuracy": 0.88,
"macro_f1": 0.8806,
"ece": 0.0808,
"brier": 0.1588,
"nll": 0.4166,
"mean_confidence": 0.9426,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.01,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0444,
"1.0": 0.12
}
},
"laya": {
"n": 100,
"accuracy": 0.31,
"macro_f1": 0.2842,
"ece": 0.4479,
"brier": 1.1179,
"nll": 3.804,
"mean_confidence": 0.7371,
"acc_at_50_coverage": 0.34,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.6719,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.64,
"0.5": 0.66,
"0.75": 0.6667,
"0.9": 0.6889,
"1.0": 0.69
}
},
"laya:english": {
"n": 100,
"accuracy": 0.12,
"macro_f1": 0.0971,
"ece": 0.5541,
"brier": 1.3544,
"nll": 6.1649,
"mean_confidence": 0.6741,
"acc_at_50_coverage": 0.12,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.8994,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.96,
"0.5": 0.88,
"0.75": 0.8667,
"0.9": 0.8667,
"1.0": 0.88
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.31,
"macro_f1": 0.2842,
"ece": 0.4479,
"brier": 1.1179,
"nll": 3.804,
"mean_confidence": 0.7371,
"acc_at_50_coverage": 0.34,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.6719,
"risk_at_coverage": {
"0.1": 0.6,
"0.25": 0.64,
"0.5": 0.66,
"0.75": 0.6667,
"0.9": 0.6889,
"1.0": 0.69
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.08,
"macro_f1": 0.0628,
"ece": 0.3566,
"brier": 1.14,
"nll": 4.1015,
"mean_confidence": 0.4366,
"acc_at_50_coverage": 0.1,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8933,
"risk_at_coverage": {
"0.1": 0.8,
"0.25": 0.92,
"0.5": 0.9,
"0.75": 0.92,
"0.9": 0.9333,
"1.0": 0.92
}
},
"jev": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.9047,
"ece": 0.0924,
"brier": 0.171,
"nll": 0.8763,
"mean_confidence": 0.8746,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.0173,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.05,
"1.0": 0.08
},
"latency_p50_ms": 247.7,
"latency_p95_ms": 641.8,
"cost_usd": 0.0027,
"input_tokens": 64388,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.33,
"kai~jev": 0.86,
"laya~jev": 0.34
}
},
"massive.te": {
"kai": {
"n": 100,
"accuracy": 0.88,
"macro_f1": 0.8625,
"ece": 0.0759,
"brier": 0.188,
"nll": 0.491,
"mean_confidence": 0.9246,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0207,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0667,
"1.0": 0.12
}
},
"laya": {
"n": 100,
"accuracy": 0.22,
"macro_f1": 0.1935,
"ece": 0.4969,
"brier": 1.1789,
"nll": 3.5847,
"mean_confidence": 0.7169,
"acc_at_50_coverage": 0.3,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.635,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.6,
"0.5": 0.7,
"0.75": 0.76,
"0.9": 0.7667,
"1.0": 0.78
}
},
"laya:english": {
"n": 100,
"accuracy": 0.09,
"macro_f1": 0.0722,
"ece": 0.6689,
"brier": 1.5148,
"nll": 7.3505,
"mean_confidence": 0.7568,
"acc_at_50_coverage": 0.06,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.9438,
"risk_at_coverage": {
"0.1": 1.0,
"0.25": 0.96,
"0.5": 0.94,
"0.75": 0.9333,
"0.9": 0.9111,
"1.0": 0.91
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.22,
"macro_f1": 0.1935,
"ece": 0.4969,
"brier": 1.1789,
"nll": 3.5847,
"mean_confidence": 0.7169,
"acc_at_50_coverage": 0.3,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.635,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.6,
"0.5": 0.7,
"0.75": 0.76,
"0.9": 0.7667,
"1.0": 0.78
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.09,
"macro_f1": 0.0615,
"ece": 0.4389,
"brier": 1.2271,
"nll": 4.4676,
"mean_confidence": 0.5289,
"acc_at_50_coverage": 0.1,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.9119,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.92,
"0.5": 0.9,
"0.75": 0.9067,
"0.9": 0.9,
"1.0": 0.91
}
},
"jev": {
"n": 100,
"accuracy": 0.91,
"macro_f1": 0.9031,
"ece": 0.0375,
"brier": 0.1252,
"nll": 0.4976,
"mean_confidence": 0.9088,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0111,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0333,
"1.0": 0.09
},
"latency_p50_ms": 240.1,
"latency_p95_ms": 353.8,
"cost_usd": 0.00269,
"input_tokens": 64020,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.23,
"kai~jev": 0.88,
"laya~jev": 0.24
}
},
"massive.th": {
"kai": {
"n": 100,
"accuracy": 0.93,
"macro_f1": 0.9301,
"ece": 0.0429,
"brier": 0.0984,
"nll": 0.2284,
"mean_confidence": 0.9729,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0054,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0111,
"1.0": 0.07
}
},
"laya": {
"n": 100,
"accuracy": 0.48,
"macro_f1": 0.4524,
"ece": 0.3563,
"brier": 0.8366,
"nll": 2.6507,
"mean_confidence": 0.8259,
"acc_at_50_coverage": 0.62,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.3897,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.32,
"0.5": 0.38,
"0.75": 0.4267,
"0.9": 0.4889,
"1.0": 0.52
}
},
"laya:english": {
"n": 100,
"accuracy": 0.08,
"macro_f1": 0.048,
"ece": 0.7179,
"brier": 1.5336,
"nll": 8.2793,
"mean_confidence": 0.7979,
"acc_at_50_coverage": 0.1,
"unanswered": 0,
"zero_prob": 0.11,
"questions": 100,
"aurc": 0.8998,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.88,
"0.5": 0.9,
"0.75": 0.8933,
"0.9": 0.9111,
"1.0": 0.92
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.48,
"macro_f1": 0.4524,
"ece": 0.3563,
"brier": 0.8366,
"nll": 2.6507,
"mean_confidence": 0.8259,
"acc_at_50_coverage": 0.62,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.3897,
"risk_at_coverage": {
"0.1": 0.3,
"0.25": 0.32,
"0.5": 0.38,
"0.75": 0.4267,
"0.9": 0.4889,
"1.0": 0.52
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.1,
"macro_f1": 0.0707,
"ece": 0.54,
"brier": 1.2942,
"nll": 4.774,
"mean_confidence": 0.64,
"acc_at_50_coverage": 0.14,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8633,
"risk_at_coverage": {
"0.1": 0.8,
"0.25": 0.8,
"0.5": 0.86,
"0.75": 0.88,
"0.9": 0.9,
"1.0": 0.9
}
},
"jev": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9415,
"ece": 0.047,
"brier": 0.0858,
"nll": 0.4225,
"mean_confidence": 0.947,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0035,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0133,
"1.0": 0.06
},
"latency_p50_ms": 255.0,
"latency_p95_ms": 337.9,
"cost_usd": 0.00268,
"input_tokens": 63742,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.49,
"kai~jev": 0.88,
"laya~jev": 0.47
}
},
"massive.tl": {
"kai": {
"n": 100,
"accuracy": 0.91,
"macro_f1": 0.9144,
"ece": 0.0598,
"brier": 0.1471,
"nll": 0.4976,
"mean_confidence": 0.9527,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0194,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.02,
"0.75": 0.0133,
"0.9": 0.0444,
"1.0": 0.09
}
},
"laya": {
"n": 100,
"accuracy": 0.3,
"macro_f1": 0.2749,
"ece": 0.4922,
"brier": 1.1279,
"nll": 4.9423,
"mean_confidence": 0.7922,
"acc_at_50_coverage": 0.46,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.447,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.28,
"0.5": 0.54,
"0.75": 0.6133,
"0.9": 0.6667,
"1.0": 0.7
}
},
"laya:english": {
"n": 100,
"accuracy": 0.29,
"macro_f1": 0.249,
"ece": 0.5121,
"brier": 1.1543,
"nll": 5.1295,
"mean_confidence": 0.8021,
"acc_at_50_coverage": 0.46,
"unanswered": 0,
"zero_prob": 0.07,
"questions": 100,
"aurc": 0.4547,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.28,
"0.5": 0.54,
"0.75": 0.6267,
"0.9": 0.6778,
"1.0": 0.71
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.35,
"macro_f1": 0.3033,
"ece": 0.4297,
"brier": 1.0109,
"nll": 3.3587,
"mean_confidence": 0.7797,
"acc_at_50_coverage": 0.54,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.3897,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.24,
"0.5": 0.46,
"0.75": 0.5733,
"0.9": 0.6222,
"1.0": 0.65
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.32,
"macro_f1": 0.2928,
"ece": 0.349,
"brier": 0.9696,
"nll": 3.4495,
"mean_confidence": 0.669,
"acc_at_50_coverage": 0.46,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.4553,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.36,
"0.5": 0.54,
"0.75": 0.6133,
"0.9": 0.6556,
"1.0": 0.68
}
},
"jev": {
"n": 100,
"accuracy": 0.85,
"macro_f1": 0.8511,
"ece": 0.0845,
"brier": 0.1962,
"nll": 1.1255,
"mean_confidence": 0.8786,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.03,
"questions": 100,
"aurc": 0.0212,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.1111,
"1.0": 0.15
},
"latency_p50_ms": 252.6,
"latency_p95_ms": 384.4,
"cost_usd": 0.00262,
"input_tokens": 62439,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.29,
"kai~jev": 0.85,
"laya~jev": 0.3
}
},
"massive.tr": {
"kai": {
"n": 100,
"accuracy": 0.89,
"macro_f1": 0.8928,
"ece": 0.0903,
"brier": 0.1769,
"nll": 0.601,
"mean_confidence": 0.9663,
"acc_at_50_coverage": 0.98,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0197,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.02,
"0.75": 0.0133,
"0.9": 0.0556,
"1.0": 0.11
}
},
"laya": {
"n": 100,
"accuracy": 0.41,
"macro_f1": 0.3977,
"ece": 0.3854,
"brier": 0.937,
"nll": 3.1366,
"mean_confidence": 0.7954,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.3757,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.24,
"0.5": 0.4,
"0.75": 0.52,
"0.9": 0.5556,
"1.0": 0.59
}
},
"laya:english": {
"n": 100,
"accuracy": 0.14,
"macro_f1": 0.0967,
"ece": 0.5874,
"brier": 1.3509,
"nll": 6.5337,
"mean_confidence": 0.7274,
"acc_at_50_coverage": 0.2,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.7108,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.68,
"0.5": 0.8,
"0.75": 0.8267,
"0.9": 0.8444,
"1.0": 0.86
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.4,
"macro_f1": 0.3846,
"ece": 0.4004,
"brier": 0.953,
"nll": 3.0366,
"mean_confidence": 0.8004,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.3915,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.24,
"0.5": 0.4,
"0.75": 0.5333,
"0.9": 0.5667,
"1.0": 0.6
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.14,
"macro_f1": 0.1107,
"ece": 0.4196,
"brier": 1.1428,
"nll": 4.3414,
"mean_confidence": 0.5596,
"acc_at_50_coverage": 0.22,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.7145,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.68,
"0.5": 0.78,
"0.75": 0.8267,
"0.9": 0.8556,
"1.0": 0.86
}
},
"jev": {
"n": 100,
"accuracy": 0.92,
"macro_f1": 0.9108,
"ece": 0.061,
"brier": 0.1492,
"nll": 0.8341,
"mean_confidence": 0.8964,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.012,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0444,
"1.0": 0.08
},
"latency_p50_ms": 258.2,
"latency_p95_ms": 371.0,
"cost_usd": 0.00262,
"input_tokens": 62421,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.43,
"kai~jev": 0.88,
"laya~jev": 0.4
}
},
"massive.ur": {
"kai": {
"n": 100,
"accuracy": 0.9,
"macro_f1": 0.8908,
"ece": 0.0527,
"brier": 0.1513,
"nll": 0.375,
"mean_confidence": 0.9427,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0097,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0444,
"1.0": 0.1
}
},
"laya": {
"n": 100,
"accuracy": 0.42,
"macro_f1": 0.3947,
"ece": 0.3355,
"brier": 0.8645,
"nll": 2.7845,
"mean_confidence": 0.7555,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.4357,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.28,
"0.5": 0.4,
"0.75": 0.48,
"0.9": 0.5333,
"1.0": 0.58
}
},
"laya:english": {
"n": 100,
"accuracy": 0.07,
"macro_f1": 0.0645,
"ece": 0.6428,
"brier": 1.4593,
"nll": 6.6436,
"mean_confidence": 0.7128,
"acc_at_50_coverage": 0.08,
"unanswered": 0,
"zero_prob": 0.05,
"questions": 100,
"aurc": 0.8953,
"risk_at_coverage": {
"0.1": 0.8,
"0.25": 0.88,
"0.5": 0.92,
"0.75": 0.9067,
"0.9": 0.9222,
"1.0": 0.93
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.42,
"macro_f1": 0.3947,
"ece": 0.3355,
"brier": 0.8645,
"nll": 2.7845,
"mean_confidence": 0.7555,
"acc_at_50_coverage": 0.6,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.4357,
"risk_at_coverage": {
"0.1": 0.4,
"0.25": 0.28,
"0.5": 0.4,
"0.75": 0.48,
"0.9": 0.5333,
"1.0": 0.58
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.11,
"macro_f1": 0.1085,
"ece": 0.4079,
"brier": 1.1741,
"nll": 4.5366,
"mean_confidence": 0.5179,
"acc_at_50_coverage": 0.16,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.8238,
"risk_at_coverage": {
"0.1": 0.8,
"0.25": 0.8,
"0.5": 0.84,
"0.75": 0.8667,
"0.9": 0.8778,
"1.0": 0.89
}
},
"jev": {
"n": 100,
"accuracy": 0.9,
"macro_f1": 0.9009,
"ece": 0.0638,
"brier": 0.1499,
"nll": 0.5641,
"mean_confidence": 0.903,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0178,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0267,
"0.9": 0.0667,
"1.0": 0.1
},
"latency_p50_ms": 258.7,
"latency_p95_ms": 384.6,
"cost_usd": 0.00266,
"input_tokens": 63380,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.42,
"kai~jev": 0.87,
"laya~jev": 0.43
}
},
"massive.vi": {
"kai": {
"n": 100,
"accuracy": 0.91,
"macro_f1": 0.8976,
"ece": 0.0697,
"brier": 0.1365,
"nll": 0.3625,
"mean_confidence": 0.9422,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.009,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0444,
"1.0": 0.09
}
},
"laya": {
"n": 100,
"accuracy": 0.33,
"macro_f1": 0.2925,
"ece": 0.467,
"brier": 1.0723,
"nll": 3.9035,
"mean_confidence": 0.7714,
"acc_at_50_coverage": 0.46,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.5551,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.4,
"0.5": 0.54,
"0.75": 0.6267,
"0.9": 0.6556,
"1.0": 0.67
}
},
"laya:english": {
"n": 100,
"accuracy": 0.06,
"macro_f1": 0.0498,
"ece": 0.69,
"brier": 1.5327,
"nll": 7.5659,
"mean_confidence": 0.75,
"acc_at_50_coverage": 0.06,
"unanswered": 0,
"zero_prob": 0.06,
"questions": 100,
"aurc": 0.886,
"risk_at_coverage": {
"0.1": 0.8,
"0.25": 0.88,
"0.5": 0.94,
"0.75": 0.9467,
"0.9": 0.9444,
"1.0": 0.94
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.34,
"macro_f1": 0.3059,
"ece": 0.4636,
"brier": 1.0617,
"nll": 3.7993,
"mean_confidence": 0.778,
"acc_at_50_coverage": 0.46,
"unanswered": 0,
"zero_prob": 0.02,
"questions": 100,
"aurc": 0.5502,
"risk_at_coverage": {
"0.1": 0.5,
"0.25": 0.4,
"0.5": 0.54,
"0.75": 0.6133,
"0.9": 0.6444,
"1.0": 0.66
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.06,
"macro_f1": 0.0474,
"ece": 0.5037,
"brier": 1.284,
"nll": 5.0598,
"mean_confidence": 0.5637,
"acc_at_50_coverage": 0.06,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.9104,
"risk_at_coverage": {
"0.1": 0.9,
"0.25": 0.92,
"0.5": 0.94,
"0.75": 0.9333,
"0.9": 0.9333,
"1.0": 0.94
}
},
"jev": {
"n": 100,
"accuracy": 0.89,
"macro_f1": 0.8817,
"ece": 0.0518,
"brier": 0.1479,
"nll": 0.5202,
"mean_confidence": 0.924,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0142,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0133,
"0.9": 0.0556,
"1.0": 0.11
},
"latency_p50_ms": 239.5,
"latency_p95_ms": 355.2,
"cost_usd": 0.00262,
"input_tokens": 62482,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.31,
"kai~jev": 0.88,
"laya~jev": 0.29
}
},
"massive.zh-CN": {
"kai": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9377,
"ece": 0.0531,
"brier": 0.0888,
"nll": 0.2722,
"mean_confidence": 0.9646,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0032,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0,
"1.0": 0.06
}
},
"laya": {
"n": 100,
"accuracy": 0.65,
"macro_f1": 0.605,
"ece": 0.2192,
"brier": 0.5484,
"nll": 1.872,
"mean_confidence": 0.859,
"acc_at_50_coverage": 0.84,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.1685,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.12,
"0.5": 0.16,
"0.75": 0.2133,
"0.9": 0.2889,
"1.0": 0.35
}
},
"laya:english": {
"n": 100,
"accuracy": 0.62,
"macro_f1": 0.6175,
"ece": 0.3198,
"brier": 0.6745,
"nll": 3.2664,
"mean_confidence": 0.9244,
"acc_at_50_coverage": 0.84,
"unanswered": 0,
"zero_prob": 0.07,
"questions": 100,
"aurc": 0.1931,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.12,
"0.5": 0.16,
"0.75": 0.2667,
"0.9": 0.3556,
"1.0": 0.38
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.65,
"macro_f1": 0.605,
"ece": 0.2192,
"brier": 0.5484,
"nll": 1.872,
"mean_confidence": 0.859,
"acc_at_50_coverage": 0.84,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.1685,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.12,
"0.5": 0.16,
"0.75": 0.2133,
"0.9": 0.2889,
"1.0": 0.35
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.6,
"macro_f1": 0.5902,
"ece": 0.2483,
"brier": 0.6039,
"nll": 1.6799,
"mean_confidence": 0.8079,
"acc_at_50_coverage": 0.84,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.1983,
"risk_at_coverage": {
"0.1": 0.1,
"0.25": 0.08,
"0.5": 0.16,
"0.75": 0.32,
"0.9": 0.3778,
"1.0": 0.4
}
},
"jev": {
"n": 100,
"accuracy": 0.94,
"macro_f1": 0.9454,
"ece": 0.0363,
"brier": 0.1018,
"nll": 0.4259,
"mean_confidence": 0.9511,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.007,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.01,
"0.9": 0.0333,
"1.0": 0.06
},
"latency_p50_ms": 263.6,
"latency_p95_ms": 517.6,
"cost_usd": 0.00262,
"input_tokens": 62289,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.65,
"kai~jev": 0.91,
"laya~jev": 0.63
}
},
"massive.zh-TW": {
"kai": {
"n": 100,
"accuracy": 0.95,
"macro_f1": 0.9489,
"ece": 0.0272,
"brier": 0.0687,
"nll": 0.2234,
"mean_confidence": 0.9712,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.0026,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0111,
"1.0": 0.05
}
},
"laya": {
"n": 100,
"accuracy": 0.61,
"macro_f1": 0.5565,
"ece": 0.2662,
"brier": 0.6563,
"nll": 2.2928,
"mean_confidence": 0.8558,
"acc_at_50_coverage": 0.78,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.2685,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.16,
"0.5": 0.22,
"0.75": 0.32,
"0.9": 0.3556,
"1.0": 0.39
}
},
"laya:english": {
"n": 100,
"accuracy": 0.46,
"macro_f1": 0.4294,
"ece": 0.4344,
"brier": 0.9242,
"nll": 4.7668,
"mean_confidence": 0.8944,
"acc_at_50_coverage": 0.72,
"unanswered": 0,
"zero_prob": 0.13,
"questions": 100,
"aurc": 0.2625,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.04,
"0.5": 0.28,
"0.75": 0.4267,
"0.9": 0.5111,
"1.0": 0.54
}
},
"laya:multilingual": {
"n": 100,
"accuracy": 0.61,
"macro_f1": 0.5565,
"ece": 0.2662,
"brier": 0.6563,
"nll": 2.2928,
"mean_confidence": 0.8558,
"acc_at_50_coverage": 0.78,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.2685,
"risk_at_coverage": {
"0.1": 0.2,
"0.25": 0.16,
"0.5": 0.22,
"0.75": 0.32,
"0.9": 0.3556,
"1.0": 0.39
}
},
"laya:typed-decisions": {
"n": 100,
"accuracy": 0.47,
"macro_f1": 0.4633,
"ece": 0.2984,
"brier": 0.7866,
"nll": 2.5075,
"mean_confidence": 0.7611,
"acc_at_50_coverage": 0.68,
"unanswered": 0,
"zero_prob": 0.0,
"questions": 100,
"aurc": 0.3109,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.2,
"0.5": 0.32,
"0.75": 0.44,
"0.9": 0.5,
"1.0": 0.53
}
},
"jev": {
"n": 100,
"accuracy": 0.93,
"macro_f1": 0.9357,
"ece": 0.0621,
"brier": 0.1145,
"nll": 0.4636,
"mean_confidence": 0.9228,
"acc_at_50_coverage": 1.0,
"unanswered": 0,
"zero_prob": 0.01,
"questions": 100,
"aurc": 0.0101,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0,
"0.5": 0.0,
"0.75": 0.0,
"0.9": 0.0444,
"1.0": 0.07
},
"latency_p50_ms": 248.6,
"latency_p95_ms": 357.2,
"cost_usd": 0.00262,
"input_tokens": 62318,
"served": {
"typesafe/jev-1.13-20260917": 100
},
"n_errors": 0
},
"agreement": {
"kai~laya": 0.59,
"kai~jev": 0.9,
"laya~jev": 0.57
}
}
},
"jev_sequential_latency": {
"n": 50,
"p50_ms": 227.3,
"p95_ms": 333.9
},
"jev_total_cost_usd": 0.2486,
"massive_summary": {
"jev": {
"languages": 51,
"n": 5099,
"unanswered": 1,
"questions": 5100,
"english": 0.92,
"non_english_macro": 0.8896,
"above_3x_random": 51,
"risk_at_coverage": {
"0.1": 0.0008,
"0.25": 0.0008,
"0.5": 0.0047,
"0.75": 0.0306,
"0.9": 0.0618,
"1.0": 0.1098
},
"macro_accuracy": 0.8902,
"macro_macro_f1": 0.8861,
"macro_ece": 0.0638,
"macro_brier": 0.1606,
"macro_nll": 0.7754,
"macro_aurc": 0.0199,
"macro_acc_at_50_coverage": 0.9953,
"macro_zero_prob": 0.0176
},
"kai": {
"languages": 51,
"n": 5100,
"unanswered": 0,
"questions": 5100,
"english": 0.94,
"non_english_macro": 0.9102,
"above_3x_random": 51,
"risk_at_coverage": {
"0.1": 0.0,
"0.25": 0.0008,
"0.5": 0.0027,
"0.75": 0.0094,
"0.9": 0.0375,
"1.0": 0.0892
},
"macro_accuracy": 0.9108,
"macro_macro_f1": 0.9025,
"macro_ece": 0.0613,
"macro_brier": 0.1367,
"macro_nll": 0.3825,
"macro_aurc": 0.0113,
"macro_acc_at_50_coverage": 0.9973,
"macro_zero_prob": 0.0
},
"laya": {
"languages": 51,
"n": 5100,
"unanswered": 0,
"questions": 5100,
"english": 0.82,
"non_english_macro": 0.3734,
"above_3x_random": 48,
"risk_at_coverage": {
"0.1": 0.3451,
"0.25": 0.4071,
"0.5": 0.4957,
"0.75": 0.559,
"0.9": 0.5919,
"1.0": 0.6178
},
"macro_accuracy": 0.3822,
"macro_macro_f1": 0.3567,
"macro_ece": 0.4184,
"macro_brier": 0.987,
"macro_nll": 3.8282,
"macro_aurc": 0.4804,
"macro_acc_at_50_coverage": 0.5043,
"macro_zero_prob": 0.0298
},
"laya:english": {
"languages": 51,
"n": 5100,
"unanswered": 0,
"questions": 5100,
"english": 0.82,
"non_english_macro": 0.215,
"above_3x_random": 23,
"risk_at_coverage": {
"0.1": 0.5392,
"0.25": 0.6212,
"0.5": 0.698,
"0.75": 0.737,
"0.9": 0.761,
"1.0": 0.7731
},
"macro_accuracy": 0.2269,
"macro_macro_f1": 0.2053,
"macro_ece": 0.5709,
"macro_brier": 1.2751,
"macro_nll": 6.3968,
"macro_aurc": 0.6716,
"macro_acc_at_50_coverage": 0.302,
"macro_zero_prob": 0.0802
},
"laya:multilingual": {
"languages": 51,
"n": 5100,
"unanswered": 0,
"questions": 5100,
"english": 0.71,
"non_english_macro": 0.3946,
"above_3x_random": 48,
"risk_at_coverage": {
"0.1": 0.3157,
"0.25": 0.3678,
"0.5": 0.458,
"0.75": 0.5336,
"0.9": 0.5715,
"1.0": 0.5992
},
"macro_accuracy": 0.4008,
"macro_macro_f1": 0.3737,
"macro_ece": 0.3911,
"macro_brier": 0.9344,
"macro_nll": 3.127,
"macro_aurc": 0.449,
"macro_acc_at_50_coverage": 0.542,
"macro_zero_prob": 0.0073
},
"laya:typed-decisions": {
"languages": 51,
"n": 5100,
"unanswered": 0,
"questions": 5100,
"english": 0.82,
"non_english_macro": 0.2104,
"above_3x_random": 24,
"risk_at_coverage": {
"0.1": 0.5275,
"0.25": 0.6204,
"0.5": 0.6941,
"0.75": 0.7422,
"0.9": 0.7634,
"1.0": 0.7776
},
"macro_accuracy": 0.2224,
"macro_macro_f1": 0.2023,
"macro_ece": 0.3989,
"macro_brier": 1.0785,
"macro_nll": 3.9626,
"macro_aurc": 0.6682,
"macro_acc_at_50_coverage": 0.3059,
"macro_zero_prob": 0.0
}
}
}