{
  "generatedAt": "2026-10-11T10:03:57.060Z",
  "seed": 20260926,
  "models": [
    "jev",
    "laya",
    "clef-flash",
    "clef",
    "microsoft-decision-1"
  ],
  "jevModel": "typesafe-ai/jev",
  "layaVersion": "0.3.20",
  "layaDevice": "mps",
  "tasks": {
    "banking77": {
      "name": "Banking77 intent (77 options)",
      "source": "https://github.com/PolyAI-LDN/task-specific-datasets",
      "license": "CC BY 4.0",
      "split": "test",
      "question": {
        "type": "choice",
        "options": 77
      },
      "jev": {
        "n": 100,
        "failed": 0,
        "accuracy": 0.79,
        "accuracy95": [
          0.7001985069803768,
          0.8583445046835422
        ],
        "meanConfidence": 0.8933999999999997,
        "ece": 0.10339999999999985,
        "predictionShare": {
          "card_payment_not_recognised": 0.02,
          "lost_or_stolen_phone": 0.01,
          "exchange_rate": 0.02,
          "top_up_failed": 0.04,
          "apple_pay_or_google_pay": 0.01,
          "passcode_forgotten": 0.01,
          "pending_cash_withdrawal": 0.03,
          "compromised_card": 0.02,
          "disposable_card_limits": 0.01,
          "exchange_charge": 0.02,
          "verify_top_up": 0.03,
          "request_refund": 0.02,
          "Refund_not_showing_up": 0.05,
          "pending_top_up": 0.01,
          "card_payment_wrong_exchange_rate": 0.01,
          "country_support": 0.01,
          "card_linking": 0.03,
          "unable_to_verify_identity": 0.03,
          "activate_my_card": 0.02,
          "change_pin": 0.06,
          "transfer_timing": 0.04,
          "transfer_not_received_by_recipient": 0.01,
          "get_disposable_virtual_card": 0.01,
          "getting_virtual_card": 0.04,
          "verify_source_of_funds": 0.02,
          "reverted_card_payment?": 0.01,
          "top_up_by_card_charge": 0.02,
          "card_not_working": 0.01,
          "verify_my_identity": 0.02,
          "direct_debit_payment_not_recognised": 0.01,
          "balance_not_updated_after_bank_transfer": 0.01,
          "declined_card_payment": 0.02,
          "pending_transfer": 0.05,
          "cash_withdrawal_charge": 0.01,
          "top_up_by_cash_or_cheque": 0.01,
          "exchange_via_app": 0.01,
          "get_physical_card": 0.01,
          "top_up_reverted": 0.01,
          "pin_blocked": 0.01,
          "atm_support": 0.03,
          "contactless_not_working": 0.01,
          "card_swallowed": 0.02,
          "transfer_into_account": 0.02,
          "top_up_limits": 0.01,
          "balance_not_updated_after_cheque_or_cash_deposit": 0.01,
          "card_about_to_expire": 0.01,
          "transfer_fee_charged": 0.03,
          "transaction_charged_twice": 0.01,
          "wrong_exchange_rate_for_cash_withdrawal": 0.01,
          "fiat_currency_support": 0.01,
          "wrong_amount_of_cash_received": 0.01,
          "lost_or_stolen_card": 0.01,
          "card_delivery_estimate": 0.01,
          "extra_charge_on_statement": 0.01
        },
        "highConfidenceShare": 0.71,
        "highConfidenceAccuracy": 0.9295774647887324,
        "latencyP50Ms": 1680,
        "latencyP95Ms": 3915,
        "meanInputTokens": 1687.44,
        "usdPer1000Calls": 0.07087248000000002,
        "reliability": [
          {
            "from": 0.4,
            "to": 0.5,
            "n": 3,
            "accuracy": 0.333,
            "confidence": 0.467
          },
          {
            "from": 0.5,
            "to": 0.6,
            "n": 8,
            "accuracy": 0.375,
            "confidence": 0.541
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 3,
            "accuracy": 0,
            "confidence": 0.637
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 7,
            "accuracy": 0.714,
            "confidence": 0.747
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 8,
            "accuracy": 0.5,
            "confidence": 0.848
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 71,
            "accuracy": 0.93,
            "confidence": 0.982
          }
        ]
      },
      "laya": {
        "n": 100,
        "failed": 0,
        "accuracy": 0.35,
        "accuracy95": [
          0.2636410465347802,
          0.4474573957080203
        ],
        "meanConfidence": 0.9012379999999998,
        "ece": 0.5568779999999999,
        "predictionShare": {
          "transaction_charged_twice": 0.05,
          "apple_pay_or_google_pay": 0.01,
          "exchange_rate": 0.04,
          "declined_transfer": 0.06,
          "top_up_reverted": 0.08,
          "passcode_forgotten": 0.04,
          "disposable_card_limits": 0.05,
          "exchange_via_app": 0.04,
          "verify_top_up": 0.03,
          "request_refund": 0.04,
          "Refund_not_showing_up": 0.02,
          "wrong_exchange_rate_for_cash_withdrawal": 0.02,
          "supported_cards_and_currencies": 0.02,
          "card_linking": 0.03,
          "verify_my_identity": 0.05,
          "activate_my_card": 0.02,
          "pin_blocked": 0.03,
          "transfer_timing": 0.08,
          "edit_personal_details": 0.02,
          "virtual_card_not_working": 0.04,
          "verify_source_of_funds": 0.02,
          "reverted_card_payment?": 0.01,
          "topping_up_by_card": 0.02,
          "direct_debit_payment_not_recognised": 0.01,
          "pending_transfer": 0.02,
          "atm_support": 0.04,
          "pending_cash_withdrawal": 0.03,
          "order_physical_card": 0.02,
          "contactless_not_working": 0.01,
          "lost_or_stolen_phone": 0.02,
          "transfer_fee_charged": 0.01,
          "declined_cash_withdrawal": 0.01,
          "extra_charge_on_statement": 0.01
        },
        "highConfidenceShare": 0.78,
        "highConfidenceAccuracy": 0.41025641025641024,
        "latencyP50Ms": 198.6,
        "latencyP95Ms": 230.2,
        "meanInputTokens": null,
        "usdPer1000Calls": null,
        "reliability": [
          {
            "from": 0.1,
            "to": 0.2,
            "n": 5,
            "accuracy": 0,
            "confidence": 0.136
          },
          {
            "from": 0.4,
            "to": 0.5,
            "n": 3,
            "accuracy": 0.333,
            "confidence": 0.472
          },
          {
            "from": 0.5,
            "to": 0.6,
            "n": 2,
            "accuracy": 0,
            "confidence": 0.57
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 3,
            "accuracy": 0,
            "confidence": 0.636
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 1,
            "accuracy": 1,
            "confidence": 0.718
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 8,
            "accuracy": 0.125,
            "confidence": 0.865
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 78,
            "accuracy": 0.41,
            "confidence": 0.992
          }
        ]
      },
      "clef-flash": {
        "n": 100,
        "failed": 0,
        "accuracy": 0.97,
        "accuracy95": [
          0.915479219195973,
          0.9897456617765851
        ],
        "meanConfidence": 0.9032909999999998,
        "ece": 0.06670900000000002,
        "predictionShare": {
          "card_payment_not_recognised": 0.01,
          "lost_or_stolen_phone": 0.01,
          "card_payment_wrong_exchange_rate": 0.02,
          "top_up_failed": 0.03,
          "apple_pay_or_google_pay": 0.01,
          "get_physical_card": 0.03,
          "pending_cash_withdrawal": 0.03,
          "compromised_card": 0.02,
          "disposable_card_limits": 0.01,
          "exchange_charge": 0.02,
          "verify_top_up": 0.03,
          "request_refund": 0.02,
          "Refund_not_showing_up": 0.05,
          "pending_top_up": 0.01,
          "country_support": 0.01,
          "card_linking": 0.03,
          "why_verify_identity": 0.02,
          "activate_my_card": 0.02,
          "change_pin": 0.04,
          "transfer_timing": 0.03,
          "transfer_not_received_by_recipient": 0.01,
          "get_disposable_virtual_card": 0.01,
          "direct_debit_payment_not_recognised": 0.02,
          "exchange_rate": 0.01,
          "getting_virtual_card": 0.04,
          "verify_source_of_funds": 0.02,
          "cancel_transfer": 0.01,
          "top_up_by_card_charge": 0.02,
          "card_not_working": 0.01,
          "verify_my_identity": 0.02,
          "balance_not_updated_after_bank_transfer": 0.03,
          "declined_transfer": 0.01,
          "pending_transfer": 0.02,
          "wrong_exchange_rate_for_cash_withdrawal": 0.02,
          "top_up_by_cash_or_cheque": 0.01,
          "top_up_reverted": 0.02,
          "exchange_via_app": 0.02,
          "order_physical_card": 0.01,
          "pin_blocked": 0.01,
          "atm_support": 0.02,
          "contactless_not_working": 0.01,
          "card_swallowed": 0.02,
          "failed_transfer": 0.01,
          "top_up_by_bank_transfer_charge": 0.02,
          "top_up_limits": 0.01,
          "balance_not_updated_after_cheque_or_cash_deposit": 0.01,
          "pending_card_payment": 0.01,
          "card_about_to_expire": 0.01,
          "transfer_fee_charged": 0.03,
          "transaction_charged_twice": 0.01,
          "declined_card_payment": 0.01,
          "unable_to_verify_identity": 0.01,
          "wrong_amount_of_cash_received": 0.01,
          "lost_or_stolen_card": 0.01,
          "declined_cash_withdrawal": 0.01,
          "card_delivery_estimate": 0.01,
          "extra_charge_on_statement": 0.01
        },
        "highConfidenceShare": 0.75,
        "highConfidenceAccuracy": 1,
        "latencyP50Ms": 881,
        "latencyP95Ms": 5274,
        "meanInputTokens": 1833.43,
        "usdPer1000Calls": 0.06967034,
        "reliability": [
          {
            "from": 0.4,
            "to": 0.5,
            "n": 2,
            "accuracy": 0.5,
            "confidence": 0.481
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 5,
            "accuracy": 0.8,
            "confidence": 0.661
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 6,
            "accuracy": 0.833,
            "confidence": 0.757
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 12,
            "accuracy": 1,
            "confidence": 0.872
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 75,
            "accuracy": 1,
            "confidence": 0.948
          }
        ]
      },
      "clef": {
        "n": 100,
        "failed": 0,
        "accuracy": 0.96,
        "accuracy95": [
          0.9016278276739262,
          0.984336949448152
        ],
        "meanConfidence": 0.9105199999999998,
        "ece": 0.06873000000000008,
        "predictionShare": {
          "card_payment_not_recognised": 0.01,
          "lost_or_stolen_phone": 0.01,
          "card_payment_wrong_exchange_rate": 0.02,
          "top_up_failed": 0.03,
          "apple_pay_or_google_pay": 0.01,
          "get_physical_card": 0.03,
          "pending_cash_withdrawal": 0.03,
          "compromised_card": 0.02,
          "disposable_card_limits": 0.01,
          "exchange_charge": 0.02,
          "verify_top_up": 0.03,
          "request_refund": 0.02,
          "Refund_not_showing_up": 0.05,
          "pending_top_up": 0.01,
          "country_support": 0.01,
          "card_linking": 0.03,
          "why_verify_identity": 0.02,
          "activate_my_card": 0.02,
          "change_pin": 0.04,
          "transfer_timing": 0.03,
          "transfer_not_received_by_recipient": 0.01,
          "get_disposable_virtual_card": 0.01,
          "direct_debit_payment_not_recognised": 0.02,
          "exchange_rate": 0.01,
          "getting_virtual_card": 0.04,
          "verify_source_of_funds": 0.02,
          "cancel_transfer": 0.01,
          "top_up_by_card_charge": 0.02,
          "card_not_working": 0.01,
          "verify_my_identity": 0.02,
          "balance_not_updated_after_bank_transfer": 0.02,
          "declined_transfer": 0.01,
          "pending_transfer": 0.04,
          "wrong_exchange_rate_for_cash_withdrawal": 0.02,
          "top_up_by_cash_or_cheque": 0.01,
          "top_up_reverted": 0.02,
          "exchange_via_app": 0.02,
          "order_physical_card": 0.01,
          "pin_blocked": 0.01,
          "atm_support": 0.02,
          "contactless_not_working": 0.01,
          "card_swallowed": 0.02,
          "top_up_by_bank_transfer_charge": 0.02,
          "top_up_limits": 0.01,
          "balance_not_updated_after_cheque_or_cash_deposit": 0.01,
          "pending_card_payment": 0.01,
          "card_about_to_expire": 0.01,
          "transfer_fee_charged": 0.03,
          "transaction_charged_twice": 0.01,
          "declined_card_payment": 0.01,
          "unable_to_verify_identity": 0.01,
          "wrong_amount_of_cash_received": 0.01,
          "lost_or_stolen_card": 0.01,
          "declined_cash_withdrawal": 0.01,
          "card_delivery_estimate": 0.01,
          "extra_charge_on_statement": 0.01
        },
        "highConfidenceShare": 0.74,
        "highConfidenceAccuracy": 1,
        "latencyP50Ms": 579,
        "latencyP95Ms": 1182,
        "meanInputTokens": 1833.43,
        "usdPer1000Calls": 0.44002319999999995,
        "reliability": [
          {
            "from": 0.4,
            "to": 0.5,
            "n": 2,
            "accuracy": 0,
            "confidence": 0.429
          },
          {
            "from": 0.5,
            "to": 0.6,
            "n": 2,
            "accuracy": 0.5,
            "confidence": 0.552
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 3,
            "accuracy": 1,
            "confidence": 0.666
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 6,
            "accuracy": 0.833,
            "confidence": 0.748
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 13,
            "accuracy": 1,
            "confidence": 0.866
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 74,
            "accuracy": 1,
            "confidence": 0.964
          }
        ]
      },
      "microsoft-decision-1": {
        "n": 100,
        "failed": 0,
        "accuracy": 0.91,
        "accuracy95": [
          0.8377362530335634,
          0.9519280048361151
        ],
        "meanConfidence": 0.8783710985438136,
        "ece": 0.04867585408443033,
        "predictionShare": {
          "card_payment_not_recognised": 0.01,
          "lost_or_stolen_phone": 0.01,
          "exchange_rate": 0.02,
          "top_up_failed": 0.03,
          "apple_pay_or_google_pay": 0.01,
          "get_physical_card": 0.03,
          "pending_cash_withdrawal": 0.03,
          "compromised_card": 0.02,
          "disposable_card_limits": 0.01,
          "exchange_charge": 0.02,
          "verify_top_up": 0.03,
          "request_refund": 0.02,
          "Refund_not_showing_up": 0.05,
          "pending_top_up": 0.01,
          "card_payment_wrong_exchange_rate": 0.01,
          "country_support": 0.01,
          "card_linking": 0.03,
          "why_verify_identity": 0.02,
          "activate_my_card": 0.02,
          "change_pin": 0.05,
          "transfer_timing": 0.03,
          "transfer_not_received_by_recipient": 0.01,
          "get_disposable_virtual_card": 0.01,
          "direct_debit_payment_not_recognised": 0.02,
          "getting_virtual_card": 0.04,
          "verify_source_of_funds": 0.02,
          "cancel_transfer": 0.01,
          "top_up_by_card_charge": 0.02,
          "card_not_working": 0.01,
          "verify_my_identity": 0.02,
          "balance_not_updated_after_bank_transfer": 0.02,
          "declined_transfer": 0.01,
          "pending_transfer": 0.04,
          "cash_withdrawal_charge": 0.01,
          "top_up_by_cash_or_cheque": 0.01,
          "top_up_reverted": 0.02,
          "exchange_via_app": 0.01,
          "pin_blocked": 0.01,
          "atm_support": 0.02,
          "contactless_not_working": 0.01,
          "card_swallowed": 0.02,
          "transfer_into_account": 0.01,
          "top_up_limits": 0.01,
          "balance_not_updated_after_cheque_or_cash_deposit": 0.01,
          "pending_card_payment": 0.01,
          "card_about_to_expire": 0.01,
          "transfer_fee_charged": 0.03,
          "transaction_charged_twice": 0.01,
          "wrong_exchange_rate_for_cash_withdrawal": 0.01,
          "fiat_currency_support": 0.01,
          "declined_card_payment": 0.01,
          "unable_to_verify_identity": 0.01,
          "wrong_amount_of_cash_received": 0.01,
          "lost_or_stolen_card": 0.01,
          "declined_cash_withdrawal": 0.01,
          "top_up_by_bank_transfer_charge": 0.01,
          "card_delivery_estimate": 0.01,
          "extra_charge_on_statement": 0.01
        },
        "highConfidenceShare": 0.66,
        "highConfidenceAccuracy": 0.9848484848484849,
        "latencyP50Ms": 735,
        "latencyP95Ms": 1409,
        "meanInputTokens": 772.2,
        "usdPer1000Calls": 0.0324324,
        "reliability": [
          {
            "from": 0.2,
            "to": 0.3,
            "n": 1,
            "accuracy": 0,
            "confidence": 0.227
          },
          {
            "from": 0.4,
            "to": 0.5,
            "n": 4,
            "accuracy": 0.75,
            "confidence": 0.453
          },
          {
            "from": 0.5,
            "to": 0.6,
            "n": 1,
            "accuracy": 0,
            "confidence": 0.523
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 11,
            "accuracy": 0.727,
            "confidence": 0.656
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 4,
            "accuracy": 1,
            "confidence": 0.763
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 13,
            "accuracy": 0.846,
            "confidence": 0.854
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 66,
            "accuracy": 0.985,
            "confidence": 0.968
          }
        ]
      }
    },
    "ag_news": {
      "name": "AG News topic (4 options)",
      "source": "https://huggingface.co/datasets/fancyzhx/ag_news",
      "license": "Non-commercial research use (see dataset card)",
      "split": "test",
      "question": {
        "type": "choice",
        "options": 4
      },
      "jev": {
        "n": 98,
        "failed": 2,
        "accuracy": 0.8367346938775511,
        "accuracy95": [
          0.751132386961264,
          0.8969328437499582
        ],
        "meanConfidence": 0.9732653061224489,
        "ece": 0.15244897959183662,
        "predictionShare": {
          "World": 0.316,
          "Sci/Tech": 0.204,
          "Business": 0.276,
          "Sports": 0.204
        },
        "highConfidenceShare": 0.9183673469387755,
        "highConfidenceAccuracy": 0.8555555555555555,
        "latencyP50Ms": 1741,
        "latencyP95Ms": 3795,
        "meanInputTokens": 431.4183673469388,
        "usdPer1000Calls": 0.01811957142857143,
        "reliability": [
          {
            "from": 0.5,
            "to": 0.6,
            "n": 1,
            "accuracy": 1,
            "confidence": 0.55
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 1,
            "accuracy": 1,
            "confidence": 0.67
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 4,
            "accuracy": 0.75,
            "confidence": 0.753
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 2,
            "accuracy": 0,
            "confidence": 0.85
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 90,
            "accuracy": 0.856,
            "confidence": 0.994
          }
        ]
      },
      "laya": {
        "n": 100,
        "failed": 0,
        "accuracy": 0.92,
        "accuracy95": [
          0.8500173312059693,
          0.9589070305141891
        ],
        "meanConfidence": 0.8780660000000003,
        "ece": 0.0419339999999999,
        "predictionShare": {
          "World": 0.28,
          "Sci/Tech": 0.29,
          "Business": 0.22,
          "Sports": 0.21
        },
        "highConfidenceShare": 0.63,
        "highConfidenceAccuracy": 0.9682539682539683,
        "latencyP50Ms": 80.7,
        "latencyP95Ms": 158.2,
        "meanInputTokens": null,
        "usdPer1000Calls": null,
        "reliability": [
          {
            "from": 0.4,
            "to": 0.5,
            "n": 4,
            "accuracy": 0.75,
            "confidence": 0.466
          },
          {
            "from": 0.5,
            "to": 0.6,
            "n": 3,
            "accuracy": 0.667,
            "confidence": 0.549
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 4,
            "accuracy": 1,
            "confidence": 0.652
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 6,
            "accuracy": 0.833,
            "confidence": 0.75
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 20,
            "accuracy": 0.85,
            "confidence": 0.85
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 63,
            "accuracy": 0.968,
            "confidence": 0.956
          }
        ]
      },
      "clef-flash": {
        "n": 100,
        "failed": 0,
        "accuracy": 0.91,
        "accuracy95": [
          0.8377362530335634,
          0.9519280048361151
        ],
        "meanConfidence": 0.8884059999999998,
        "ece": 0.04074799999999989,
        "predictionShare": {
          "World": 0.31,
          "Sci/Tech": 0.28,
          "Business": 0.2,
          "Sports": 0.21
        },
        "highConfidenceShare": 0.64,
        "highConfidenceAccuracy": 0.9375,
        "latencyP50Ms": 546,
        "latencyP95Ms": 2074,
        "meanInputTokens": 257.15,
        "usdPer1000Calls": 0.0097717,
        "reliability": [
          {
            "from": 0.5,
            "to": 0.6,
            "n": 3,
            "accuracy": 0.667,
            "confidence": 0.538
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 4,
            "accuracy": 0.5,
            "confidence": 0.646
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 7,
            "accuracy": 1,
            "confidence": 0.751
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 22,
            "accuracy": 0.909,
            "confidence": 0.864
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 64,
            "accuracy": 0.938,
            "confidence": 0.943
          }
        ]
      },
      "clef": {
        "n": 100,
        "failed": 0,
        "accuracy": 0.91,
        "accuracy95": [
          0.8377362530335634,
          0.9519280048361151
        ],
        "meanConfidence": 0.9254290000000001,
        "ece": 0.021467000000000215,
        "predictionShare": {
          "World": 0.31,
          "Sci/Tech": 0.29,
          "Business": 0.19,
          "Sports": 0.21
        },
        "highConfidenceShare": 0.82,
        "highConfidenceAccuracy": 0.9512195121951219,
        "latencyP50Ms": 585,
        "latencyP95Ms": 1146,
        "meanInputTokens": 257.15,
        "usdPer1000Calls": 0.06171599999999999,
        "reliability": [
          {
            "from": 0.3,
            "to": 0.4,
            "n": 1,
            "accuracy": 0,
            "confidence": 0.368
          },
          {
            "from": 0.5,
            "to": 0.6,
            "n": 2,
            "accuracy": 0.5,
            "confidence": 0.588
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 3,
            "accuracy": 0.667,
            "confidence": 0.663
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 5,
            "accuracy": 0.8,
            "confidence": 0.742
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 7,
            "accuracy": 0.857,
            "confidence": 0.857
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 82,
            "accuracy": 0.951,
            "confidence": 0.967
          }
        ]
      },
      "microsoft-decision-1": {
        "n": 100,
        "failed": 0,
        "accuracy": 0.83,
        "accuracy95": [
          0.7445181876635703,
          0.8910652394022684
        ],
        "meanConfidence": 0.9531907920473276,
        "ece": 0.13912869082649218,
        "predictionShare": {
          "World": 0.31,
          "Sci/Tech": 0.19,
          "Business": 0.29,
          "Sports": 0.21
        },
        "highConfidenceShare": 0.86,
        "highConfidenceAccuracy": 0.872093023255814,
        "latencyP50Ms": 699,
        "latencyP95Ms": 1359,
        "meanInputTokens": 139.44,
        "usdPer1000Calls": 0.00585648,
        "reliability": [
          {
            "from": 0.4,
            "to": 0.5,
            "n": 1,
            "accuracy": 0,
            "confidence": 0.499
          },
          {
            "from": 0.5,
            "to": 0.6,
            "n": 2,
            "accuracy": 0.5,
            "confidence": 0.561
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 7,
            "accuracy": 0.857,
            "confidence": 0.743
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 4,
            "accuracy": 0.25,
            "confidence": 0.851
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 86,
            "accuracy": 0.872,
            "confidence": 0.989
          }
        ]
      }
    },
    "emotion": {
      "name": "Emotion (6 options)",
      "source": "https://huggingface.co/datasets/dair-ai/emotion",
      "license": "Educational and research use (see dataset card)",
      "split": "test",
      "question": {
        "type": "choice",
        "options": 6
      },
      "jev": {
        "n": 49,
        "failed": 0,
        "accuracy": 0.6530612244897959,
        "accuracy95": [
          0.5131093521943979,
          0.7707579039068555
        ],
        "meanConfidence": 0.8497959183673471,
        "ece": 0.22530612244897968,
        "predictionShare": {
          "sadness": 0.408,
          "joy": 0.245,
          "fear": 0.204,
          "anger": 0.082,
          "surprise": 0.041,
          "love": 0.02
        },
        "highConfidenceShare": 0.5102040816326531,
        "highConfidenceAccuracy": 0.76,
        "latencyP50Ms": 1778,
        "latencyP95Ms": 4225,
        "meanInputTokens": 445.16326530612247,
        "usdPer1000Calls": 0.018696857142857144,
        "reliability": [
          {
            "from": 0.3,
            "to": 0.4,
            "n": 1,
            "accuracy": 1,
            "confidence": 0.37
          },
          {
            "from": 0.4,
            "to": 0.5,
            "n": 2,
            "accuracy": 0.5,
            "confidence": 0.465
          },
          {
            "from": 0.5,
            "to": 0.6,
            "n": 3,
            "accuracy": 0.333,
            "confidence": 0.567
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 2,
            "accuracy": 0.5,
            "confidence": 0.63
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 5,
            "accuracy": 0.6,
            "confidence": 0.742
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 11,
            "accuracy": 0.545,
            "confidence": 0.84
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 25,
            "accuracy": 0.76,
            "confidence": 0.977
          }
        ]
      },
      "laya": {
        "n": 49,
        "failed": 0,
        "accuracy": 0.5714285714285714,
        "accuracy95": [
          0.4327024751701555,
          0.6997689110104294
        ],
        "meanConfidence": 0.8224530612244897,
        "ece": 0.3004408163265305,
        "predictionShare": {
          "sadness": 0.367,
          "joy": 0.367,
          "anger": 0.061,
          "fear": 0.143,
          "love": 0.061
        },
        "highConfidenceShare": 0.5102040816326531,
        "highConfidenceAccuracy": 0.64,
        "latencyP50Ms": 81.8,
        "latencyP95Ms": 108.9,
        "meanInputTokens": null,
        "usdPer1000Calls": null,
        "reliability": [
          {
            "from": 0.3,
            "to": 0.4,
            "n": 1,
            "accuracy": 0,
            "confidence": 0.391
          },
          {
            "from": 0.4,
            "to": 0.5,
            "n": 4,
            "accuracy": 0,
            "confidence": 0.468
          },
          {
            "from": 0.5,
            "to": 0.6,
            "n": 3,
            "accuracy": 0.667,
            "confidence": 0.53
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 5,
            "accuracy": 0.4,
            "confidence": 0.641
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 7,
            "accuracy": 0.857,
            "confidence": 0.743
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 4,
            "accuracy": 0.5,
            "confidence": 0.873
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 25,
            "accuracy": 0.64,
            "confidence": 0.982
          }
        ]
      },
      "clef-flash": {
        "n": 49,
        "failed": 0,
        "accuracy": 0.5714285714285714,
        "accuracy95": [
          0.4327024751701555,
          0.6997689110104294
        ],
        "meanConfidence": 0.7794857142857141,
        "ece": 0.23355102040816328,
        "predictionShare": {
          "sadness": 0.449,
          "joy": 0.224,
          "fear": 0.102,
          "anger": 0.122,
          "surprise": 0.041,
          "love": 0.061
        },
        "highConfidenceShare": 0.2857142857142857,
        "highConfidenceAccuracy": 0.7857142857142857,
        "latencyP50Ms": 510,
        "latencyP95Ms": 2235,
        "meanInputTokens": 276.83673469387753,
        "usdPer1000Calls": 0.010519795918367344,
        "reliability": [
          {
            "from": 0.3,
            "to": 0.4,
            "n": 1,
            "accuracy": 1,
            "confidence": 0.375
          },
          {
            "from": 0.4,
            "to": 0.5,
            "n": 3,
            "accuracy": 0,
            "confidence": 0.475
          },
          {
            "from": 0.5,
            "to": 0.6,
            "n": 7,
            "accuracy": 0.286,
            "confidence": 0.552
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 4,
            "accuracy": 0.25,
            "confidence": 0.646
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 5,
            "accuracy": 0.6,
            "confidence": 0.742
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 15,
            "accuracy": 0.667,
            "confidence": 0.864
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 14,
            "accuracy": 0.786,
            "confidence": 0.948
          }
        ]
      },
      "clef": {
        "n": 49,
        "failed": 0,
        "accuracy": 0.5510204081632653,
        "accuracy95": [
          0.41314866799678107,
          0.6814737507036366
        ],
        "meanConfidence": 0.769342857142857,
        "ece": 0.21832244897959188,
        "predictionShare": {
          "sadness": 0.449,
          "joy": 0.224,
          "anger": 0.143,
          "surprise": 0.041,
          "fear": 0.102,
          "love": 0.041
        },
        "highConfidenceShare": 0.3469387755102041,
        "highConfidenceAccuracy": 0.7647058823529411,
        "latencyP50Ms": 572,
        "latencyP95Ms": 1033,
        "meanInputTokens": 276.83673469387753,
        "usdPer1000Calls": 0.06644081632653061,
        "reliability": [
          {
            "from": 0.3,
            "to": 0.4,
            "n": 1,
            "accuracy": 0,
            "confidence": 0.35
          },
          {
            "from": 0.4,
            "to": 0.5,
            "n": 5,
            "accuracy": 0,
            "confidence": 0.435
          },
          {
            "from": 0.5,
            "to": 0.6,
            "n": 6,
            "accuracy": 0.167,
            "confidence": 0.55
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 4,
            "accuracy": 0.25,
            "confidence": 0.62
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 5,
            "accuracy": 0.6,
            "confidence": 0.738
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 11,
            "accuracy": 0.818,
            "confidence": 0.868
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 17,
            "accuracy": 0.765,
            "confidence": 0.95
          }
        ]
      },
      "microsoft-decision-1": {
        "n": 49,
        "failed": 0,
        "accuracy": 0.6326530612244898,
        "accuracy95": [
          0.49266811386889864,
          0.7533501747521878
        ],
        "meanConfidence": 0.7424982259868806,
        "ece": 0.21111413018114972,
        "predictionShare": {
          "sadness": 0.347,
          "joy": 0.245,
          "fear": 0.184,
          "anger": 0.102,
          "surprise": 0.061,
          "love": 0.061
        },
        "highConfidenceShare": 0.24489795918367346,
        "highConfidenceAccuracy": 0.75,
        "latencyP50Ms": 720,
        "latencyP95Ms": 1347,
        "meanInputTokens": 135.83673469387756,
        "usdPer1000Calls": 0.005705142857142857,
        "reliability": [
          {
            "from": 0.2,
            "to": 0.3,
            "n": 1,
            "accuracy": 1,
            "confidence": 0.257
          },
          {
            "from": 0.3,
            "to": 0.4,
            "n": 3,
            "accuracy": 0,
            "confidence": 0.355
          },
          {
            "from": 0.4,
            "to": 0.5,
            "n": 3,
            "accuracy": 1,
            "confidence": 0.434
          },
          {
            "from": 0.5,
            "to": 0.6,
            "n": 5,
            "accuracy": 0.2,
            "confidence": 0.536
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 3,
            "accuracy": 0.667,
            "confidence": 0.653
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 12,
            "accuracy": 0.583,
            "confidence": 0.756
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 10,
            "accuracy": 0.8,
            "confidence": 0.863
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 12,
            "accuracy": 0.75,
            "confidence": 0.951
          }
        ]
      }
    },
    "sst2": {
      "name": "SST-2 sentiment (yes/no)",
      "source": "https://huggingface.co/datasets/stanfordnlp/sst2",
      "license": "See dataset card",
      "split": "validation",
      "question": {
        "type": "boolean"
      },
      "laya": {
        "n": 200,
        "failed": 0,
        "accuracy": 0.48,
        "accuracy95": [
          0.4117904407948983,
          0.5489633994418346
        ],
        "meanConfidence": 0.9725279999999994,
        "ece": 0.5013979999999996,
        "predictionShare": {
          "negative": 0.975,
          "positive": 0.025
        },
        "highConfidenceShare": 0.92,
        "highConfidenceAccuracy": 0.4945652173913043,
        "latencyP50Ms": 116.8,
        "latencyP95Ms": 600.7,
        "meanInputTokens": null,
        "usdPer1000Calls": null,
        "reliability": [
          {
            "from": 0.5,
            "to": 0.6,
            "n": 2,
            "accuracy": 1,
            "confidence": 0.557
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 4,
            "accuracy": 0.25,
            "confidence": 0.65
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 4,
            "accuracy": 0,
            "confidence": 0.736
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 6,
            "accuracy": 0.333,
            "confidence": 0.849
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 184,
            "accuracy": 0.495,
            "confidence": 0.993
          }
        ]
      },
      "clef-flash": {
        "n": 200,
        "failed": 0,
        "accuracy": 0.96,
        "accuracy95": [
          0.9230670617058012,
          0.9795946128493436
        ],
        "meanConfidence": 0.9524995000000002,
        "ece": 0.028740500000000297,
        "predictionShare": {
          "negative": 0.475,
          "positive": 0.525
        },
        "highConfidenceShare": 0.94,
        "highConfidenceAccuracy": 0.9840425531914894,
        "latencyP50Ms": 397,
        "latencyP95Ms": 1459,
        "meanInputTokens": 169.615,
        "usdPer1000Calls": 0.006445370000000001,
        "reliability": [
          {
            "from": 0.5,
            "to": 0.6,
            "n": 2,
            "accuracy": 0.5,
            "confidence": 0.527
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 1,
            "accuracy": 0,
            "confidence": 0.682
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 3,
            "accuracy": 0.667,
            "confidence": 0.749
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 6,
            "accuracy": 0.667,
            "confidence": 0.857
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 188,
            "accuracy": 0.984,
            "confidence": 0.965
          }
        ]
      },
      "clef": {
        "n": 200,
        "failed": 0,
        "accuracy": 0.965,
        "accuracy95": [
          0.9295285391333187,
          0.9829446753626429
        ],
        "meanConfidence": 0.943518,
        "ece": 0.033638000000000154,
        "predictionShare": {
          "negative": 0.47,
          "positive": 0.53
        },
        "highConfidenceShare": 0.87,
        "highConfidenceAccuracy": 0.9827586206896551,
        "latencyP50Ms": 530,
        "latencyP95Ms": 1285,
        "meanInputTokens": 169.615,
        "usdPer1000Calls": 0.0407076,
        "reliability": [
          {
            "from": 0.5,
            "to": 0.6,
            "n": 4,
            "accuracy": 0.75,
            "confidence": 0.539
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 7,
            "accuracy": 0.571,
            "confidence": 0.745
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 15,
            "accuracy": 1,
            "confidence": 0.86
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 174,
            "accuracy": 0.983,
            "confidence": 0.968
          }
        ]
      },
      "microsoft-decision-1": {
        "n": 200,
        "failed": 0,
        "accuracy": 0.97,
        "accuracy95": [
          0.9361048802675267,
          0.9861798741692518
        ],
        "meanConfidence": 0.9745210820515401,
        "ece": 0.009497933326935718,
        "predictionShare": {
          "negative": 0.465,
          "positive": 0.535
        },
        "highConfidenceShare": 0.93,
        "highConfidenceAccuracy": 0.9838709677419355,
        "latencyP50Ms": 716,
        "latencyP95Ms": 1480,
        "meanInputTokens": 77.615,
        "usdPer1000Calls": 0.00325983,
        "reliability": [
          {
            "from": 0.5,
            "to": 0.6,
            "n": 1,
            "accuracy": 1,
            "confidence": 0.531
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 1,
            "accuracy": 0,
            "confidence": 0.651
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 4,
            "accuracy": 0.75,
            "confidence": 0.759
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 8,
            "accuracy": 0.875,
            "confidence": 0.871
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 186,
            "accuracy": 0.984,
            "confidence": 0.988
          }
        ]
      }
    },
    "sst2_choice": {
      "name": "SST-2 sentiment as a choice (2 options)",
      "source": "https://huggingface.co/datasets/stanfordnlp/sst2",
      "license": "See dataset card",
      "split": "validation",
      "question": {
        "type": "choice",
        "options": 2
      },
      "laya": {
        "n": 200,
        "failed": 0,
        "accuracy": 0.93,
        "accuracy95": [
          0.8859442178331628,
          0.9578482170770813
        ],
        "meanConfidence": 0.8872254999999997,
        "ece": 0.08814550000000002,
        "predictionShare": {
          "negative": 0.435,
          "positive": 0.565
        },
        "highConfidenceShare": 0.67,
        "highConfidenceAccuracy": 0.9850746268656716,
        "latencyP50Ms": 55.6,
        "latencyP95Ms": 69.5,
        "meanInputTokens": null,
        "usdPer1000Calls": null,
        "reliability": [
          {
            "from": 0.5,
            "to": 0.6,
            "n": 12,
            "accuracy": 0.167,
            "confidence": 0.545
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 12,
            "accuracy": 0.917,
            "confidence": 0.652
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 13,
            "accuracy": 1,
            "confidence": 0.751
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 29,
            "accuracy": 0.966,
            "confidence": 0.859
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 134,
            "accuracy": 0.985,
            "confidence": 0.958
          }
        ]
      },
      "clef-flash": {
        "n": 200,
        "failed": 0,
        "accuracy": 0.97,
        "accuracy95": [
          0.9361048802675267,
          0.9861798741692518
        ],
        "meanConfidence": 0.9615554999999997,
        "ece": 0.02088950000000042,
        "predictionShare": {
          "negative": 0.455,
          "positive": 0.545
        },
        "highConfidenceShare": 0.945,
        "highConfidenceAccuracy": 0.9841269841269841,
        "latencyP50Ms": 423,
        "latencyP95Ms": 1058,
        "meanInputTokens": 167.615,
        "usdPer1000Calls": 0.00636937,
        "reliability": [
          {
            "from": 0.5,
            "to": 0.6,
            "n": 2,
            "accuracy": 1,
            "confidence": 0.515
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 3,
            "accuracy": 0.667,
            "confidence": 0.609
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 1,
            "accuracy": 1,
            "confidence": 0.778
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 5,
            "accuracy": 0.6,
            "confidence": 0.849
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 189,
            "accuracy": 0.984,
            "confidence": 0.976
          }
        ]
      },
      "clef": {
        "n": 200,
        "failed": 0,
        "accuracy": 0.965,
        "accuracy95": [
          0.9295285391333187,
          0.9829446753626429
        ],
        "meanConfidence": 0.9609354999999996,
        "ece": 0.007686500000000072,
        "predictionShare": {
          "negative": 0.47,
          "positive": 0.53
        },
        "highConfidenceShare": 0.915,
        "highConfidenceAccuracy": 0.9836065573770492,
        "latencyP50Ms": 562,
        "latencyP95Ms": 1256,
        "meanInputTokens": 167.615,
        "usdPer1000Calls": 0.0402276,
        "reliability": [
          {
            "from": 0.6,
            "to": 0.7,
            "n": 3,
            "accuracy": 0.667,
            "confidence": 0.668
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 10,
            "accuracy": 0.7,
            "confidence": 0.736
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 4,
            "accuracy": 1,
            "confidence": 0.882
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 183,
            "accuracy": 0.984,
            "confidence": 0.98
          }
        ]
      },
      "microsoft-decision-1": {
        "n": 200,
        "failed": 0,
        "accuracy": 0.97,
        "accuracy95": [
          0.9361048802675267,
          0.9861798741692518
        ],
        "meanConfidence": 0.9681180530273291,
        "ece": 0.007993319366377708,
        "predictionShare": {
          "negative": 0.465,
          "positive": 0.535
        },
        "highConfidenceShare": 0.895,
        "highConfidenceAccuracy": 0.9888268156424581,
        "latencyP50Ms": 688,
        "latencyP95Ms": 1381,
        "meanInputTokens": 75.615,
        "usdPer1000Calls": 0.00317583,
        "reliability": [
          {
            "from": 0.5,
            "to": 0.6,
            "n": 2,
            "accuracy": 0.5,
            "confidence": 0.577
          },
          {
            "from": 0.6,
            "to": 0.7,
            "n": 2,
            "accuracy": 1,
            "confidence": 0.651
          },
          {
            "from": 0.7,
            "to": 0.8,
            "n": 7,
            "accuracy": 0.714,
            "confidence": 0.764
          },
          {
            "from": 0.8,
            "to": 0.9,
            "n": 10,
            "accuracy": 0.9,
            "confidence": 0.871
          },
          {
            "from": 0.9,
            "to": 1,
            "n": 179,
            "accuracy": 0.989,
            "confidence": 0.989
          }
        ]
      }
    }
  }
}