{
 "schema": 1,
 "date": "2026-10-14",
 "edition": "2026.10",
 "machine": "rtx-5090",
 "device": "cuda",
 "suite": "public-benchmark",
 "servers": {
  "ollaya": "0.13.0"
 },
 "protocol": {
  "benchmark": "Bespoke Labs' public decision benchmark: 13 subsets of 11 human-labeled public datasets, 3,880 questions",
  "source": "https://github.com/bespokelabsai/nimble/blob/main/docs/PUBLIC_BENCHMARKS.md",
  "runner": "Bespoke's own evaluate_public_jev, through convert/ollaya_convert/bench_public.py",
  "client": "one request (one question) at a time over HTTP to /v1/systemone, timed by the client",
  "accuracy": "macro: the mean of the 13 subsets; a rejected request counts as wrong",
  "ece": "ten-bin expected calibration error of the top probability; pooled over all 3,880 questions"
 },
 "note": "Ollaya Bench edition 2026.10.",
 "results": [
  {
   "server": "ollaya",
   "model": "jeb:27b",
   "digest": "da59bd8b0f90123b7ca0de07f2a325aa860b8ab97814ba36226f11463186f561",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7883,
   "macro_accuracy_ci": [
    0.7747,
    0.8019
   ],
   "micro_accuracy": 0.7969,
   "macro_ece": 0.0616,
   "pooled_ece": 0.0235,
   "brier": 0.2832,
   "median_ms": 97.8,
   "p95_ms": 238.5,
   "by_type": {
    "choice": 0.8474,
    "noul": 0.8237,
    "score": 0.6309
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.8147,
     "ece": 0.0654
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8771,
     "ece": 0.0323
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8686,
     "ece": 0.0549
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.89,
     "ece": 0.0474
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7625,
     "ece": 0.1302
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.828,
     "ece": 0.0715
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.9164,
     "ece": 0.0275
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.85,
     "ece": 0.0595
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.788,
     "ece": 0.0587
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.4578,
     "ece": 0.061
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.5667,
     "ece": 0.0312
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8681,
     "ece": 0.1086
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.76,
     "ece": 0.0529
    }
   }
  },
  {
   "server": "ollaya",
   "model": "winnow:12b",
   "digest": "8d72fee0e5f92a625cdb85ae4f6abdc024c9e6f124a0ca67e64522c3f24c47ce",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7726,
   "macro_accuracy_ci": [
    0.7599,
    0.7848
   ],
   "micro_accuracy": 0.7778,
   "macro_ece": 0.1458,
   "pooled_ece": 0.1415,
   "brier": 0.3623,
   "median_ms": 56.8,
   "p95_ms": 108.6,
   "by_type": {
    "choice": 0.8,
    "noul": 0.8535,
    "score": 0.5922
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7546,
     "ece": 0.1907
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8657,
     "ece": 0.1059
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8514,
     "ece": 0.1019
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8733,
     "ece": 0.0898
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8763,
     "ece": 0.085
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.832,
     "ece": 0.1102
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8161,
     "ece": 0.1348
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.87,
     "ece": 0.0818
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.816,
     "ece": 0.1456
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.4056,
     "ece": 0.3895
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.4958,
     "ece": 0.1986
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.875,
     "ece": 0.087
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.712,
     "ece": 0.1749
    }
   }
  },
  {
   "server": "ollaya",
   "model": "clef:flash",
   "digest": "acda6b87d74dc55e04916220fcfc643ca2d409ca22ec5e9d18bf53b91604845d",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7718,
   "macro_accuracy_ci": [
    0.7589,
    0.7856
   ],
   "micro_accuracy": 0.782,
   "macro_ece": 0.0724,
   "pooled_ece": 0.0249,
   "brier": 0.306,
   "median_ms": 283.5,
   "p95_ms": 556.9,
   "by_type": {
    "choice": 0.8356,
    "noul": 0.8359,
    "score": 0.5587
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7679,
     "ece": 0.0604
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.9086,
     "ece": 0.0308
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.9029,
     "ece": 0.041
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8567,
     "ece": 0.0625
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8227,
     "ece": 0.0421
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.812,
     "ece": 0.0394
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8227,
     "ece": 0.039
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.9,
     "ece": 0.0553
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.788,
     "ece": 0.0657
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.3815,
     "ece": 0.1564
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.475,
     "ece": 0.089
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8194,
     "ece": 0.2078
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.776,
     "ece": 0.0524
    }
   }
  },
  {
   "server": "ollaya",
   "model": "cygnet:12b",
   "digest": "81160d85e13acfda209f78ba38a3e1d7ef0c80727820b022bbd9c878476e93cc",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7676,
   "macro_accuracy_ci": [
    0.755,
    0.7805
   ],
   "micro_accuracy": 0.7729,
   "macro_ece": 0.1238,
   "pooled_ece": 0.1034,
   "brier": 0.3519,
   "median_ms": 47.5,
   "p95_ms": 99.5,
   "by_type": {
    "choice": 0.8099,
    "noul": 0.8388,
    "score": 0.5786
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7479,
     "ece": 0.1816
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8571,
     "ece": 0.0627
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8714,
     "ece": 0.0517
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8667,
     "ece": 0.0645
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8629,
     "ece": 0.0664
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.832,
     "ece": 0.0866
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8328,
     "ece": 0.0975
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8167,
     "ece": 0.1112
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.816,
     "ece": 0.1386
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.4137,
     "ece": 0.3304
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.4542,
     "ece": 0.1638
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8681,
     "ece": 0.1071
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.74,
     "ece": 0.147
    }
   }
  },
  {
   "server": "ollaya",
   "model": "jeb:9b",
   "digest": "dfe0411a8d060f1fa64ff2539b1e4b44a061c11cce4d5d96a82a7077b01d8b9e",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7663,
   "macro_accuracy_ci": [
    0.7534,
    0.7788
   ],
   "micro_accuracy": 0.7709,
   "macro_ece": 0.0694,
   "pooled_ece": 0.0389,
   "brier": 0.3113,
   "median_ms": 33.7,
   "p95_ms": 74.3,
   "by_type": {
    "choice": 0.8044,
    "noul": 0.8256,
    "score": 0.6038
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7546,
     "ece": 0.0894
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8514,
     "ece": 0.0316
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8457,
     "ece": 0.0367
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8567,
     "ece": 0.0765
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7993,
     "ece": 0.1011
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.828,
     "ece": 0.0608
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8462,
     "ece": 0.0405
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.86,
     "ece": 0.0292
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.784,
     "ece": 0.1031
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.4337,
     "ece": 0.114
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.4958,
     "ece": 0.0425
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8819,
     "ece": 0.1243
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.724,
     "ece": 0.0527
    }
   }
  },
  {
   "server": "ollaya",
   "model": "jeb:4b",
   "digest": "4aba27707f6f4ba6f116710a0d05743220e5a5819b8826a9bd32385d0d9df2ff",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7589,
   "macro_accuracy_ci": [
    0.7461,
    0.7731
   ],
   "micro_accuracy": 0.766,
   "macro_ece": 0.0734,
   "pooled_ece": 0.0481,
   "brier": 0.3286,
   "median_ms": 27.1,
   "p95_ms": 55.3,
   "by_type": {
    "choice": 0.8025,
    "noul": 0.8213,
    "score": 0.5821
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7596,
     "ece": 0.1161
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8571,
     "ece": 0.0344
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8314,
     "ece": 0.0324
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.85,
     "ece": 0.076
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7625,
     "ece": 0.1333
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.832,
     "ece": 0.0671
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8763,
     "ece": 0.0252
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.87,
     "ece": 0.0631
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.792,
     "ece": 0.0608
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.3936,
     "ece": 0.1001
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.5125,
     "ece": 0.0349
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8403,
     "ece": 0.1196
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.688,
     "ece": 0.0912
    }
   }
  },
  {
   "server": "ollaya",
   "model": "decider:4b",
   "digest": "920dea4d1851680da7129db890b97542e824a3709f2ccb09b7f35ce6b4ff4669",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7565,
   "macro_accuracy_ci": [
    0.7436,
    0.7696
   ],
   "micro_accuracy": 0.7649,
   "macro_ece": 0.0887,
   "pooled_ece": 0.0427,
   "brier": 0.3279,
   "median_ms": 179.4,
   "p95_ms": 1294.7,
   "by_type": {
    "choice": 0.8169,
    "noul": 0.8203,
    "score": 0.5495
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7412,
     "ece": 0.1472
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8629,
     "ece": 0.041
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8429,
     "ece": 0.0278
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8667,
     "ece": 0.0359
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7893,
     "ece": 0.0429
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.768,
     "ece": 0.1219
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.9298,
     "ece": 0.0138
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8733,
     "ece": 0.0565
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.804,
     "ece": 0.0467
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.4498,
     "ece": 0.1422
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.3792,
     "ece": 0.063
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8194,
     "ece": 0.3299
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.708,
     "ece": 0.0844
    }
   }
  },
  {
   "server": "ollaya",
   "model": "kev:9b",
   "digest": "c9130c1a1e629c13497c833300395b56c7e347ecd3558596a2c2e1ce41ce7eda",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7526,
   "macro_accuracy_ci": [
    0.7394,
    0.7659
   ],
   "micro_accuracy": 0.7588,
   "macro_ece": 0.1059,
   "pooled_ece": 0.0583,
   "brier": 0.3563,
   "median_ms": 227.4,
   "p95_ms": 503.5,
   "by_type": {
    "choice": 0.816,
    "noul": 0.8073,
    "score": 0.5558
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7346,
     "ece": 0.1563
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8457,
     "ece": 0.0484
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8371,
     "ece": 0.0662
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8467,
     "ece": 0.0724
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7425,
     "ece": 0.1521
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.804,
     "ece": 0.1216
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.9064,
     "ece": 0.0375
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8433,
     "ece": 0.0558
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.8,
     "ece": 0.0844
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.3855,
     "ece": 0.1351
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.4625,
     "ece": 0.0714
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8194,
     "ece": 0.3039
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.756,
     "ece": 0.0717
    }
   }
  },
  {
   "server": "ollaya",
   "model": "d1:3b",
   "digest": "d92cc117d0d9ef8624b4408ef4823b4b8efef403fee9f7d54629cb037a2b58da",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7507,
   "macro_accuracy_ci": [
    0.7372,
    0.7632
   ],
   "micro_accuracy": 0.7554,
   "macro_ece": 0.1054,
   "pooled_ece": 0.0625,
   "brier": 0.3335,
   "median_ms": 14.8,
   "p95_ms": 28.7,
   "by_type": {
    "choice": 0.7905,
    "noul": 0.8755,
    "score": 0.4764
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.6828,
     "ece": 0.1527
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.86,
     "ece": 0.0454
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8429,
     "ece": 0.0318
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.85,
     "ece": 0.0579
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8896,
     "ece": 0.0458
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.904,
     "ece": 0.0233
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8428,
     "ece": 0.0548
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.91,
     "ece": 0.0415
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.824,
     "ece": 0.0743
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.4418,
     "ece": 0.1065
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.1542,
     "ece": 0.4566
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8333,
     "ece": 0.2078
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.724,
     "ece": 0.0716
    }
   }
  },
  {
   "server": "ollaya",
   "model": "nimble:9b",
   "digest": "fb7e7f742e34636e853f91c467b9463e3155dbb08cabc5c49906d52a5750487d",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7476,
   "macro_accuracy_ci": [
    0.7346,
    0.7611
   ],
   "micro_accuracy": 0.7562,
   "macro_ece": 0.0996,
   "pooled_ece": 0.0225,
   "brier": 0.3389,
   "median_ms": 327.7,
   "p95_ms": 676.6,
   "by_type": {
    "choice": 0.7997,
    "noul": 0.8263,
    "score": 0.5295
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7679,
     "ece": 0.065
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.84,
     "ece": 0.0826
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8086,
     "ece": 0.1012
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8533,
     "ece": 0.0304
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8227,
     "ece": 0.0238
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.8,
     "ece": 0.0762
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8261,
     "ece": 0.0679
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8433,
     "ece": 0.1198
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.812,
     "ece": 0.0499
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.3414,
     "ece": 0.159
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.4,
     "ece": 0.081
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8472,
     "ece": 0.3789
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.756,
     "ece": 0.0589
    }
   }
  },
  {
   "server": "ollaya",
   "model": "kev:4b",
   "digest": "c0e732655dc9d799dd21ba7f2f9a445fbd50ee83578e96ccfbf691c1be9dde53",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7448,
   "macro_accuracy_ci": [
    0.7308,
    0.758
   ],
   "micro_accuracy": 0.7528,
   "macro_ece": 0.1129,
   "pooled_ece": 0.062,
   "brier": 0.3595,
   "median_ms": 187,
   "p95_ms": 400.5,
   "by_type": {
    "choice": 0.8153,
    "noul": 0.8161,
    "score": 0.5082
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7212,
     "ece": 0.1563
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8629,
     "ece": 0.1604
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.86,
     "ece": 0.2062
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8533,
     "ece": 0.0639
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7826,
     "ece": 0.0622
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.816,
     "ece": 0.0608
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.9164,
     "ece": 0.0429
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8167,
     "ece": 0.102
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.812,
     "ece": 0.0751
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.3414,
     "ece": 0.0989
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.3708,
     "ece": 0.0372
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8125,
     "ece": 0.3662
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.716,
     "ece": 0.0359
    }
   }
  },
  {
   "server": "ollaya",
   "model": "jeeves:9b",
   "digest": "0f4a1ff8396c0718c432603d96c0a8eb975a0d3bab83bd374bd35e36dfc5512d",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7441,
   "macro_accuracy_ci": [
    0.7315,
    0.7575
   ],
   "micro_accuracy": 0.7438,
   "macro_ece": 0.0832,
   "pooled_ece": 0.0488,
   "brier": 0.3601,
   "median_ms": 298.6,
   "p95_ms": 576.4,
   "by_type": {
    "choice": 0.7971,
    "noul": 0.8163,
    "score": 0.5353
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.6327,
     "ece": 0.2378
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8657,
     "ece": 0.0376
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8514,
     "ece": 0.0365
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.87,
     "ece": 0.0554
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8294,
     "ece": 0.0175
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.82,
     "ece": 0.0907
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.9197,
     "ece": 0.0344
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.79,
     "ece": 0.069
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.772,
     "ece": 0.041
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.3614,
     "ece": 0.0924
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.4042,
     "ece": 0.0754
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8403,
     "ece": 0.2333
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.716,
     "ece": 0.0609
    }
   }
  },
  {
   "server": "ollaya",
   "model": "credence:e4b",
   "digest": "ab2f84b8484681a88d465b7771f43804886e7bf4d529a731d4b8ce24bea57b69",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7414,
   "macro_accuracy_ci": [
    0.728,
    0.7542
   ],
   "micro_accuracy": 0.743,
   "macro_ece": 0.094,
   "pooled_ece": 0.0545,
   "brier": 0.3722,
   "median_ms": 41.2,
   "p95_ms": 74.1,
   "by_type": {
    "choice": 0.7743,
    "noul": 0.811,
    "score": 0.5704
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.6861,
     "ece": 0.159
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8657,
     "ece": 0.0365
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8257,
     "ece": 0.0452
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8133,
     "ece": 0.1058
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8595,
     "ece": 0.0305
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.792,
     "ece": 0.0295
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.806,
     "ece": 0.0232
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.77,
     "ece": 0.0431
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.82,
     "ece": 0.0966
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.3695,
     "ece": 0.2299
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.4875,
     "ece": 0.0583
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8542,
     "ece": 0.2463
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.688,
     "ece": 0.1176
    }
   }
  },
  {
   "server": "ollaya",
   "model": "credence:e4b-calibrated",
   "digest": "e9d941a002f0bba8438d09f8c8fe9bdad8a830dab0ea8cba13cd758e1fafb565",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7368,
   "macro_accuracy_ci": [
    0.7227,
    0.751
   ],
   "micro_accuracy": 0.7389,
   "macro_ece": 0.0901,
   "pooled_ece": 0.0378,
   "brier": 0.3693,
   "median_ms": 41.1,
   "p95_ms": 73.9,
   "by_type": {
    "choice": 0.768,
    "noul": 0.8102,
    "score": 0.5628
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.6861,
     "ece": 0.1456
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.86,
     "ece": 0.0342
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8057,
     "ece": 0.0585
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.81,
     "ece": 0.1069
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8328,
     "ece": 0.0518
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.796,
     "ece": 0.0556
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.796,
     "ece": 0.0689
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8,
     "ece": 0.0346
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.812,
     "ece": 0.0771
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.3855,
     "ece": 0.1867
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.4625,
     "ece": 0.0298
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8403,
     "ece": 0.2511
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.692,
     "ece": 0.0702
    }
   }
  },
  {
   "server": "ollaya",
   "model": "winnow:e4b",
   "digest": "dd4bf88aa50bebb02e7a26fbebed14682befd037e3e89789ee22b9793f82d226",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7343,
   "macro_accuracy_ci": [
    0.7206,
    0.7475
   ],
   "micro_accuracy": 0.7371,
   "macro_ece": 0.1023,
   "pooled_ece": 0.0714,
   "brier": 0.3802,
   "median_ms": 41.3,
   "p95_ms": 74.3,
   "by_type": {
    "choice": 0.7762,
    "noul": 0.798,
    "score": 0.5582
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.6962,
     "ece": 0.1696
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8514,
     "ece": 0.0237
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8086,
     "ece": 0.0505
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8167,
     "ece": 0.0998
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8161,
     "ece": 0.0595
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.808,
     "ece": 0.0437
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8328,
     "ece": 0.0387
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.7533,
     "ece": 0.048
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.796,
     "ece": 0.0872
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.3775,
     "ece": 0.3159
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.45,
     "ece": 0.0731
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8472,
     "ece": 0.2233
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.692,
     "ece": 0.0974
    }
   }
  },
  {
   "server": "ollaya",
   "model": "decider:2b",
   "digest": "5507d6f80e9a2c82d7dd4c0eb6cffc03916669a4ce5f0fe506ab5e434cf9a62a",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7033,
   "macro_accuracy_ci": [
    0.6895,
    0.7174
   ],
   "micro_accuracy": 0.7106,
   "macro_ece": 0.1075,
   "pooled_ece": 0.0872,
   "brier": 0.4021,
   "median_ms": 100.9,
   "p95_ms": 628.6,
   "by_type": {
    "choice": 0.7669,
    "noul": 0.7732,
    "score": 0.4808
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.6361,
     "ece": 0.213
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.82,
     "ece": 0.0687
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.7943,
     "ece": 0.0752
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8033,
     "ece": 0.0836
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7826,
     "ece": 0.0687
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.72,
     "ece": 0.1475
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8562,
     "ece": 0.0431
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.84,
     "ece": 0.0376
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.72,
     "ece": 0.1476
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.4257,
     "ece": 0.0904
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.35,
     "ece": 0.0486
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.6667,
     "ece": 0.3138
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.728,
     "ece": 0.0591
    }
   }
  },
  {
   "server": "ollaya",
   "model": "snap:2b",
   "digest": "f356b13c94415f1237ab94d752b0de1705ed98f38ee389549264dfa53532f171",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.7017,
   "macro_accuracy_ci": [
    0.6878,
    0.7158
   ],
   "micro_accuracy": 0.7082,
   "macro_ece": 0.1033,
   "pooled_ece": 0.049,
   "brier": 0.3986,
   "median_ms": 19.8,
   "p95_ms": 36.1,
   "by_type": {
    "choice": 0.7066,
    "noul": 0.814,
    "score": 0.5064
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7245,
     "ece": 0.0532
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.7771,
     "ece": 0.095
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.6943,
     "ece": 0.0999
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.7933,
     "ece": 0.0707
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7625,
     "ece": 0.0428
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.888,
     "ece": 0.1274
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7692,
     "ece": 0.1286
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.89,
     "ece": 0.0945
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.736,
     "ece": 0.0869
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.3735,
     "ece": 0.1174
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.3125,
     "ece": 0.0599
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8333,
     "ece": 0.2856
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.568,
     "ece": 0.0816
    }
   }
  },
  {
   "server": "ollaya",
   "model": "snap:2b-q4_k_m",
   "digest": "a19db7952873b56351a1a1e820fa571c65a8f50b4e9862addbf0dda063661036",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.6907,
   "macro_accuracy_ci": [
    0.6767,
    0.7038
   ],
   "micro_accuracy": 0.6987,
   "macro_ece": 0.1135,
   "pooled_ece": 0.0435,
   "brier": 0.415,
   "median_ms": 19.8,
   "p95_ms": 36.2,
   "by_type": {
    "choice": 0.6902,
    "noul": 0.8055,
    "score": 0.5001
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7179,
     "ece": 0.0583
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.7571,
     "ece": 0.114
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.66,
     "ece": 0.1163
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.7733,
     "ece": 0.058
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.796,
     "ece": 0.1009
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.864,
     "ece": 0.1046
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.796,
     "ece": 0.1586
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.89,
     "ece": 0.0494
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.704,
     "ece": 0.112
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.3655,
     "ece": 0.1166
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.35,
     "ece": 0.0642
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.7847,
     "ece": 0.3017
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.52,
     "ece": 0.1202
    }
   }
  },
  {
   "server": "ollaya",
   "model": "decider:0.8b",
   "digest": "97a22ac9c6c42eebb762ddb5eb47f0542b5e0788cf1cea31cee6205d5bd1064a",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.6687,
   "macro_accuracy_ci": [
    0.654,
    0.6814
   ],
   "micro_accuracy": 0.6799,
   "macro_ece": 0.1237,
   "pooled_ece": 0.0612,
   "brier": 0.421,
   "median_ms": 93.5,
   "p95_ms": 518.5,
   "by_type": {
    "choice": 0.7449,
    "noul": 0.7187,
    "score": 0.4585
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.6778,
     "ece": 0.1041
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8171,
     "ece": 0.0707
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.7829,
     "ece": 0.0651
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.7933,
     "ece": 0.0326
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7224,
     "ece": 0.085
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.592,
     "ece": 0.1471
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8227,
     "ece": 0.0313
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.87,
     "ece": 0.0308
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.616,
     "ece": 0.1876
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.3976,
     "ece": 0.0895
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.1375,
     "ece": 0.2462
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8403,
     "ece": 0.44
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.624,
     "ece": 0.0775
    }
   }
  },
  {
   "server": "ollaya",
   "model": "jevk5:4b",
   "digest": "58b5d6516410070d5f165d77357ed7bfefe5ffedde63e6ae75ab5101d4fe0221",
   "records": 3880,
   "errors": 700,
   "macro_accuracy": 0.6205,
   "macro_accuracy_ci": [
    0.6071,
    0.6322
   ],
   "micro_accuracy": 0.599,
   "macro_ece": 0.0702,
   "pooled_ece": 0.04,
   "brier": 0.373,
   "median_ms": 24.9,
   "p95_ms": 58.7,
   "by_type": {
    "choice": 0.4539,
    "noul": 0.8175,
    "score": 0.5698
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.6828,
     "ece": 0.0886
    },
    "massive-en-US": {
     "n": 350,
     "errors": 350,
     "accuracy": 0,
     "ece": null
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 350,
     "accuracy": 0,
     "ece": null
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8467,
     "ece": 0.03
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7993,
     "ece": 0.0277
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.808,
     "ece": 0.0481
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8428,
     "ece": 0.0489
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8333,
     "ece": 0.0376
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.8,
     "ece": 0.1031
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.4177,
     "ece": 0.1289
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.4375,
     "ece": 0.0566
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.8542,
     "ece": 0.1034
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.744,
     "ece": 0.0989
    }
   }
  },
  {
   "server": "ollaya",
   "model": "kev:0.8b",
   "digest": "8a0d4094df504ca7329bc999dabbb89c15ad5221f55728acde0ff61c2b3909f2",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.6106,
   "macro_accuracy_ci": [
    0.5968,
    0.6252
   ],
   "micro_accuracy": 0.6314,
   "macro_ece": 0.128,
   "pooled_ece": 0.0345,
   "brier": 0.4651,
   "median_ms": 103.4,
   "p95_ms": 200.5,
   "by_type": {
    "choice": 0.6637,
    "noul": 0.7288,
    "score": 0.3251
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.6628,
     "ece": 0.0735
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.7486,
     "ece": 0.207
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.66,
     "ece": 0.1629
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.7933,
     "ece": 0.0597
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.6622,
     "ece": 0.0162
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.66,
     "ece": 0.0669
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7993,
     "ece": 0.0515
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8367,
     "ece": 0.137
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.692,
     "ece": 0.0893
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.2892,
     "ece": 0.1086
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.1375,
     "ece": 0.2744
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.5486,
     "ece": 0.2256
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.448,
     "ece": 0.1915
    }
   }
  },
  {
   "server": "ollaya",
   "model": "laya:typed-decisions",
   "digest": "6d17e5fbcbb8215a74f2efd0dcc0ffdbd472e9b25ccb07867f9f146a791673ba",
   "records": 3880,
   "errors": 31,
   "macro_accuracy": 0.5935,
   "macro_accuracy_ci": [
    0.5785,
    0.6084
   ],
   "micro_accuracy": 0.6235,
   "macro_ece": 0.1247,
   "pooled_ece": 0.0219,
   "brier": 0.4755,
   "median_ms": 21.3,
   "p95_ms": 38.8,
   "by_type": {
    "choice": 0.6392,
    "noul": 0.697,
    "score": 0.3447
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.783,
     "ece": 0.0921
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.6571,
     "ece": 0.1113
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.3743,
     "ece": 0.2116
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8233,
     "ece": 0.0408
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.699,
     "ece": 0.0407
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.54,
     "ece": 0.1769
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8696,
     "ece": 0.0869
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.9267,
     "ece": 0.2429
    },
    "aegis2": {
     "n": 250,
     "errors": 1,
     "accuracy": 0.496,
     "ece": 0.155
    },
    "helpsteer2": {
     "n": 249,
     "errors": 30,
     "accuracy": 0.3534,
     "ece": 0.0322
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.2292,
     "ece": 0.1417
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.4514,
     "ece": 0.1924
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.512,
     "ece": 0.0967
    }
   }
  },
  {
   "server": "ollaya",
   "model": "decision:eos",
   "digest": "fc119d7067ecd029a5316442cf42a394236fff8758ab296f3d25a1285a34e047",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.5905,
   "macro_accuracy_ci": [
    0.5757,
    0.6048
   ],
   "micro_accuracy": 0.6206,
   "macro_ece": 0.1158,
   "pooled_ece": 0.087,
   "brier": 0.478,
   "median_ms": 93.1,
   "p95_ms": 183,
   "by_type": {
    "choice": 0.6674,
    "noul": 0.7139,
    "score": 0.2565
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.6244,
     "ece": 0.1027
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8171,
     "ece": 0.0573
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.7543,
     "ece": 0.0582
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.6833,
     "ece": 0.1294
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8729,
     "ece": 0.0609
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.672,
     "ece": 0.2006
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7692,
     "ece": 0.0477
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.6533,
     "ece": 0.1218
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.688,
     "ece": 0.048
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.257,
     "ece": 0.2262
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.2208,
     "ece": 0.1048
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.2917,
     "ece": 0.1242
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.372,
     "ece": 0.2242
    }
   }
  },
  {
   "server": "ollaya",
   "model": "laya:multilingual",
   "digest": "2840506e1f978aeb696a6f84af43ecc7fca95754cb63a67bafe6132532d384eb",
   "records": 3880,
   "errors": 27,
   "macro_accuracy": 0.5787,
   "macro_accuracy_ci": [
    0.5639,
    0.5933
   ],
   "micro_accuracy": 0.6121,
   "macro_ece": 0.178,
   "pooled_ece": 0.1548,
   "brier": 0.5365,
   "median_ms": 15.2,
   "p95_ms": 17.5,
   "by_type": {
    "choice": 0.639,
    "noul": 0.7287,
    "score": 0.2282
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7629,
     "ece": 0.0748
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.6229,
     "ece": 0.124
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.4429,
     "ece": 0.2858
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.7533,
     "ece": 0.1551
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.602,
     "ece": 0.3104
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.768,
     "ece": 0.0761
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8462,
     "ece": 0.0962
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.92,
     "ece": 0.0679
    },
    "aegis2": {
     "n": 250,
     "errors": 1,
     "accuracy": 0.6,
     "ece": 0.258
    },
    "helpsteer2": {
     "n": 249,
     "errors": 26,
     "accuracy": 0.261,
     "ece": 0.1265
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.125,
     "ece": 0.3258
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.2986,
     "ece": 0.14
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.52,
     "ece": 0.2738
    }
   }
  },
  {
   "server": "ollaya",
   "model": "decima:base",
   "digest": "56ad25422fccba646359e24655a82fc836a07b760cdc121aced5a56f222c25f1",
   "records": 3880,
   "errors": 301,
   "macro_accuracy": 0.5503,
   "macro_accuracy_ci": [
    0.5369,
    0.5639
   ],
   "micro_accuracy": 0.599,
   "macro_ece": 0.1033,
   "pooled_ece": 0.0525,
   "brier": 0.441,
   "median_ms": 25.1,
   "p95_ms": 26.5,
   "by_type": {
    "choice": 0.7256,
    "noul": 0.6244,
    "score": 0.1344
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.6845,
     "ece": 0.0946
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8257,
     "ece": 0.0327
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8057,
     "ece": 0.0607
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.6533,
     "ece": 0.0481
    },
    "squad2": {
     "n": 299,
     "errors": 19,
     "accuracy": 0.5853,
     "ece": 0.0376
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.616,
     "ece": 0.1021
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8562,
     "ece": 0.0612
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8033,
     "ece": 0.0183
    },
    "aegis2": {
     "n": 250,
     "errors": 9,
     "accuracy": 0.464,
     "ece": 0.2468
    },
    "helpsteer2": {
     "n": 249,
     "errors": 107,
     "accuracy": 0.1727,
     "ece": 0.1912
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 105,
     "accuracy": 0.1542,
     "ece": 0.1453
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 48,
     "accuracy": 0.0764,
     "ece": 0.2542
    },
    "pubmedqa": {
     "n": 250,
     "errors": 13,
     "accuracy": 0.456,
     "ece": 0.0505
    }
   }
  },
  {
   "server": "ollaya",
   "model": "laya",
   "digest": null,
   "records": 3880,
   "errors": 452,
   "macro_accuracy": 0.5348,
   "macro_accuracy_ci": [
    0.5216,
    0.5472
   ],
   "micro_accuracy": 0.5884,
   "macro_ece": 0.1744,
   "pooled_ece": 0.1032,
   "brier": 0.4757,
   "median_ms": 17.1,
   "p95_ms": 36.6,
   "by_type": {
    "choice": 0.6583,
    "noul": 0.68,
    "score": 0.0869
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7896,
     "ece": 0.129
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.6857,
     "ece": 0.1375
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.44,
     "ece": 0.2586
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8367,
     "ece": 0.0807
    },
    "squad2": {
     "n": 299,
     "errors": 19,
     "accuracy": 0.6187,
     "ece": 0.1819
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.58,
     "ece": 0.2107
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8763,
     "ece": 0.0579
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.9367,
     "ece": 0.1714
    },
    "aegis2": {
     "n": 250,
     "errors": 7,
     "accuracy": 0.428,
     "ece": 0.3411
    },
    "helpsteer2": {
     "n": 249,
     "errors": 144,
     "accuracy": 0.1566,
     "ece": 0.0894
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 170,
     "accuracy": 0.0625,
     "ece": 0.1428
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 101,
     "accuracy": 0.0417,
     "ece": 0.1958
    },
    "pubmedqa": {
     "n": 250,
     "errors": 11,
     "accuracy": 0.5,
     "ece": 0.27
    }
   }
  },
  {
   "server": "ollaya",
   "model": "decima:agent",
   "digest": "6b5f4883a460936fd5dd3b427fcd57b6711b668495bbdc5c960f5fbb62532c83",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.5326,
   "macro_accuracy_ci": [
    0.5178,
    0.5469
   ],
   "micro_accuracy": 0.5778,
   "macro_ece": 0.2562,
   "pooled_ece": 0.2107,
   "brier": 0.6223,
   "median_ms": 27.3,
   "p95_ms": 30.7,
   "by_type": {
    "choice": 0.6834,
    "noul": 0.6069,
    "score": 0.1572
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.6311,
     "ece": 0.2379
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8371,
     "ece": 0.0708
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8057,
     "ece": 0.0453
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.6,
     "ece": 0.2166
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.5719,
     "ece": 0.209
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.512,
     "ece": 0.45
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7793,
     "ece": 0.1764
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.7667,
     "ece": 0.0668
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.584,
     "ece": 0.0855
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.1687,
     "ece": 0.5296
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.2542,
     "ece": 0.3952
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.0486,
     "ece": 0.6223
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.364,
     "ece": 0.2256
    }
   }
  },
  {
   "server": "ollaya",
   "model": "laya:en",
   "digest": "c305a9276531a47000bf93559d2c94f1ed6cbb67055c9151084682dad7655e9d",
   "records": 3880,
   "errors": 453,
   "macro_accuracy": 0.5321,
   "macro_accuracy_ci": [
    0.5198,
    0.5447
   ],
   "micro_accuracy": 0.5853,
   "macro_ece": 0.1765,
   "pooled_ece": 0.1063,
   "brier": 0.48,
   "median_ms": 17.8,
   "p95_ms": 20.5,
   "by_type": {
    "choice": 0.652,
    "noul": 0.6792,
    "score": 0.0869
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7896,
     "ece": 0.129
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.6857,
     "ece": 0.1375
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.4086,
     "ece": 0.2871
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8367,
     "ece": 0.0807
    },
    "squad2": {
     "n": 299,
     "errors": 19,
     "accuracy": 0.6187,
     "ece": 0.1819
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.576,
     "ece": 0.2128
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.8763,
     "ece": 0.0579
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.9367,
     "ece": 0.1714
    },
    "aegis2": {
     "n": 250,
     "errors": 8,
     "accuracy": 0.428,
     "ece": 0.3384
    },
    "helpsteer2": {
     "n": 249,
     "errors": 144,
     "accuracy": 0.1566,
     "ece": 0.0894
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 170,
     "accuracy": 0.0625,
     "ece": 0.1428
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 101,
     "accuracy": 0.0417,
     "ece": 0.1958
    },
    "pubmedqa": {
     "n": 250,
     "errors": 11,
     "accuracy": 0.5,
     "ece": 0.27
    }
   }
  },
  {
   "server": "ollaya",
   "model": "von:1.1",
   "digest": "25606bfd644593e63b4c63b8219d4fd0085d11634d6fa4701268d77cf804f689",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.4853,
   "macro_accuracy_ci": [
    0.4707,
    0.4991
   ],
   "micro_accuracy": 0.526,
   "macro_ece": 0.1815,
   "pooled_ece": 0.0979,
   "brier": 0.6248,
   "median_ms": 16.9,
   "p95_ms": 21,
   "by_type": {
    "choice": 0.5164,
    "noul": 0.6365,
    "score": 0.1815
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.7613,
     "ece": 0.2366
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.6914,
     "ece": 0.2641
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.2657,
     "ece": 0.5944
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.7167,
     "ece": 0.1275
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.5017,
     "ece": 0.1688
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.576,
     "ece": 0.0393
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.3278,
     "ece": 0.1417
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.76,
     "ece": 0.144
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.628,
     "ece": 0.0563
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.1446,
     "ece": 0.1433
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.3167,
     "ece": 0.0643
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.0833,
     "ece": 0.2673
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.536,
     "ece": 0.1112
    }
   }
  },
  {
   "server": "ollaya",
   "model": "decima:small",
   "digest": "d0cdad8529bb48df9511afb3f16e97254bceb40d9ec38a6dd4064930e2352db3",
   "records": 3880,
   "errors": 445,
   "macro_accuracy": 0.4622,
   "macro_accuracy_ci": [
    0.449,
    0.4755
   ],
   "micro_accuracy": 0.5075,
   "macro_ece": 0.1326,
   "pooled_ece": 0.0646,
   "brier": 0.5162,
   "median_ms": 14.4,
   "p95_ms": 16.6,
   "by_type": {
    "choice": 0.6143,
    "noul": 0.5411,
    "score": 0.0771
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.5726,
     "ece": 0.0838
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.8057,
     "ece": 0.0815
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.7143,
     "ece": 0.0522
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.4733,
     "ece": 0.0594
    },
    "squad2": {
     "n": 299,
     "errors": 28,
     "accuracy": 0.4415,
     "ece": 0.0579
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.516,
     "ece": 0.0949
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7191,
     "ece": 0.0356
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8667,
     "ece": 0.1075
    },
    "aegis2": {
     "n": 250,
     "errors": 12,
     "accuracy": 0.408,
     "ece": 0.4428
    },
    "helpsteer2": {
     "n": 249,
     "errors": 130,
     "accuracy": 0.0924,
     "ece": 0.1908
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 159,
     "accuracy": 0.0625,
     "ece": 0.1921
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 96,
     "accuracy": 0.0764,
     "ece": 0.1339
    },
    "pubmedqa": {
     "n": 250,
     "errors": 20,
     "accuracy": 0.26,
     "ece": 0.1916
    }
   }
  },
  {
   "server": "ollaya",
   "model": "nli:deberta-v3-large",
   "digest": "b33142f79a0338ab5398c7be66c18f58483c07aaec8262dc1f314aa02cbee205",
   "records": 3880,
   "errors": 326,
   "macro_accuracy": 0.4583,
   "macro_accuracy_ci": [
    0.4447,
    0.4731
   ],
   "micro_accuracy": 0.4665,
   "macro_ece": 0.2092,
   "pooled_ece": 0.1956,
   "brier": 0.6352,
   "median_ms": 14.9,
   "p95_ms": 62.2,
   "by_type": {
    "choice": 0.4433,
    "noul": 0.6639,
    "score": 0.1405
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.2604,
     "ece": 0.1569
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.6,
     "ece": 0.41
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.5457,
     "ece": 0.4059
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.7467,
     "ece": 0.0586
    },
    "squad2": {
     "n": 299,
     "errors": 19,
     "accuracy": 0.4749,
     "ece": 0.4351
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.596,
     "ece": 0.0537
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.3144,
     "ece": 0.2782
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.89,
     "ece": 0.0672
    },
    "aegis2": {
     "n": 250,
     "errors": 7,
     "accuracy": 0.612,
     "ece": 0.2001
    },
    "helpsteer2": {
     "n": 249,
     "errors": 124,
     "accuracy": 0.1687,
     "ece": 0.1401
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 115,
     "accuracy": 0.225,
     "ece": 0.1025
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 53,
     "accuracy": 0.0278,
     "ece": 0.3603
    },
    "pubmedqa": {
     "n": 250,
     "errors": 8,
     "accuracy": 0.496,
     "ece": 0.0508
    }
   }
  },
  {
   "server": "ollaya",
   "model": "gliclass:large",
   "digest": "8c4c3acece90b83f44da3694f0aedcb90503de37aff702995e18dc317a63e5a6",
   "records": 3880,
   "errors": 28,
   "macro_accuracy": 0.4336,
   "macro_accuracy_ci": [
    0.4193,
    0.4482
   ],
   "micro_accuracy": 0.4263,
   "macro_ece": 0.3018,
   "pooled_ece": 0.2838,
   "brier": 0.8105,
   "median_ms": 15.1,
   "p95_ms": 26.1,
   "by_type": {
    "choice": 0.3658,
    "noul": 0.6001,
    "score": 0.2689
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.1202,
     "ece": 0.6138
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.5743,
     "ece": 0.1701
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.4943,
     "ece": 0.1679
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.58,
     "ece": 0.3773
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.505,
     "ece": 0.3296
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.544,
     "ece": 0.077
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.3244,
     "ece": 0.5065
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8033,
     "ece": 0.045
    },
    "aegis2": {
     "n": 250,
     "errors": 1,
     "accuracy": 0.568,
     "ece": 0.3016
    },
    "helpsteer2": {
     "n": 249,
     "errors": 27,
     "accuracy": 0.2651,
     "ece": 0.0413
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.4583,
     "ece": 0.272
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.0833,
     "ece": 0.6608
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.316,
     "ece": 0.3602
    }
   }
  },
  {
   "server": "ollaya",
   "model": "nli:modernbert-large",
   "digest": "0c984ae95685d511953341fd75e5cf49e31bae5110574dba78999c5f2267cd7f",
   "records": 3880,
   "errors": 306,
   "macro_accuracy": 0.4232,
   "macro_accuracy_ci": [
    0.4094,
    0.4369
   ],
   "micro_accuracy": 0.4531,
   "macro_ece": 0.1798,
   "pooled_ece": 0.1111,
   "brier": 0.6554,
   "median_ms": 16.8,
   "p95_ms": 48.5,
   "by_type": {
    "choice": 0.3977,
    "noul": 0.6323,
    "score": 0.1171
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.5793,
     "ece": 0.036
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.3829,
     "ece": 0.1428
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.2457,
     "ece": 0.0704
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.6933,
     "ece": 0.0755
    },
    "squad2": {
     "n": 299,
     "errors": 19,
     "accuracy": 0.4682,
     "ece": 0.4518
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.544,
     "ece": 0.2485
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.4448,
     "ece": 0.0253
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.88,
     "ece": 0.0682
    },
    "aegis2": {
     "n": 250,
     "errors": 7,
     "accuracy": 0.576,
     "ece": 0.1872
    },
    "helpsteer2": {
     "n": 249,
     "errors": 109,
     "accuracy": 0.1888,
     "ece": 0.072
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 115,
     "accuracy": 0.1208,
     "ece": 0.1896
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 51,
     "accuracy": 0.0417,
     "ece": 0.4589
    },
    "pubmedqa": {
     "n": 250,
     "errors": 5,
     "accuracy": 0.336,
     "ece": 0.3115
    }
   }
  },
  {
   "server": "ollaya",
   "model": "arbiter:4b",
   "digest": "2896b8aecfca38e37f22e75e78dca0a6955eba6cfd8dc5b24b637eb124b8cff8",
   "records": 3880,
   "errors": 1333,
   "macro_accuracy": 0.4198,
   "macro_accuracy_ci": [
    0.4089,
    0.4317
   ],
   "micro_accuracy": 0.443,
   "macro_ece": 0.1393,
   "pooled_ece": 0.1372,
   "brier": 0.4755,
   "median_ms": 61.6,
   "p95_ms": 101.7,
   "by_type": {
    "choice": 0.381,
    "noul": 0.7105,
    "score": 0
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.611,
     "ece": 0.2265
    },
    "massive-en-US": {
     "n": 350,
     "errors": 350,
     "accuracy": 0,
     "ece": null
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 350,
     "accuracy": 0,
     "ece": null
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.7967,
     "ece": 0.0635
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7492,
     "ece": 0.0327
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.724,
     "ece": 0.0932
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.7258,
     "ece": 0.1056
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.5867,
     "ece": 0.2476
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.696,
     "ece": 0.1761
    },
    "helpsteer2": {
     "n": 249,
     "errors": 249,
     "accuracy": 0,
     "ece": null
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 240,
     "accuracy": 0,
     "ece": null
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 144,
     "accuracy": 0,
     "ece": null
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.568,
     "ece": 0.1692
    }
   }
  },
  {
   "server": "ollaya",
   "model": "clm:8b",
   "digest": "3f89a66972f1398210d2f41b94667665132aa51c2d313237c3fc70a49c275cb5",
   "records": 3880,
   "errors": 0,
   "macro_accuracy": 0.3767,
   "macro_accuracy_ci": [
    0.363,
    0.3899
   ],
   "micro_accuracy": 0.3861,
   "macro_ece": 0.3858,
   "pooled_ece": 0.3725,
   "brier": 0.916,
   "median_ms": 71.3,
   "p95_ms": 224.3,
   "by_type": {
    "choice": 0.3186,
    "noul": 0.5737,
    "score": 0.1453
   },
   "per_subset": {
    "vitaminc-dev": {
     "n": 599,
     "errors": 0,
     "accuracy": 0.3873,
     "ece": 0.585
    },
    "massive-en-US": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.08,
     "ece": 0.6593
    },
    "massive-de-DE": {
     "n": 350,
     "errors": 0,
     "accuracy": 0.06,
     "ece": 0.5369
    },
    "boolq": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.5633,
     "ece": 0.1408
    },
    "squad2": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.5217,
     "ece": 0.3301
    },
    "paws": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.516,
     "ece": 0.4809
    },
    "multinli": {
     "n": 299,
     "errors": 0,
     "accuracy": 0.6656,
     "ece": 0.071
    },
    "civil_comments": {
     "n": 300,
     "errors": 0,
     "accuracy": 0.8433,
     "ece": 0.0563
    },
    "aegis2": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.424,
     "ece": 0.3905
    },
    "helpsteer2": {
     "n": 249,
     "errors": 0,
     "accuracy": 0.1928,
     "ece": 0.2605
    },
    "summeval-relevance": {
     "n": 240,
     "errors": 0,
     "accuracy": 0.2292,
     "ece": 0.3622
    },
    "summeval-consistency": {
     "n": 144,
     "errors": 0,
     "accuracy": 0.0139,
     "ece": 0.8379
    },
    "pubmedqa": {
     "n": 250,
     "errors": 0,
     "accuracy": 0.4,
     "ece": 0.3039
    }
   }
  }
 ]
}
