{
  "schemaVersion": 2,
  "retrievedAt": "2026-10-13",
  "primarySource": "https://artificialanalysis.ai/evaluations/frontiercode",
  "sourceHtmlSha256": "791988e690cd0e94596067bc213b4a039fb035315a5e9dd340ac15cba2bfb146",
  "collection": "Manually retrieved Artificial Analysis public pages on 13 October 2026. Score values use one consistent 22-variant source record set. Definitions were checked against each evaluation page. No automatic collection or runtime API.",
  "benchmarks": [
    {
      "key": "frontierCode",
      "name": "FrontierCode 1.1",
      "url": "https://artificialanalysis.ai/evaluations/frontiercode",
      "exposure": "private",
      "unit": "percent",
      "sourceField": "frontierCode"
    },
    {
      "key": "briefcaseRubric",
      "name": "AA-Briefcase v1.1 \u00b7 rubric",
      "url": "https://artificialanalysis.ai/evaluations/aa-briefcase",
      "exposure": "private",
      "unit": "percent",
      "sourceField": "briefcaseBreakdown.rubricPassRate"
    },
    {
      "key": "automation",
      "name": "AutomationBench-AA",
      "url": "https://artificialanalysis.ai/evaluations/automationbench-aa",
      "exposure": "private",
      "unit": "percent",
      "sourceField": "automationBenchPartialScore"
    },
    {
      "key": "analyst",
      "name": "AA-AnalystAgent \u00b7 pass^5",
      "url": "https://artificialanalysis.ai/evaluations/aa-analyst-agent",
      "exposure": "private",
      "unit": "percent",
      "sourceField": "analystAgent"
    },
    {
      "key": "omniscienceAccuracy",
      "name": "AA-Omniscience \u00b7 accuracy",
      "url": "https://artificialanalysis.ai/evaluations/omniscience",
      "exposure": "private",
      "unit": "percent",
      "sourceField": "omniscienceBreakdown.accuracy",
      "note": "Accuracy over all questions, not the -100\u2013100 hallucination-adjusted Omniscience Index."
    },
    {
      "key": "harveyLab",
      "name": "Harvey LAB-AA v1.1",
      "url": "https://artificialanalysis.ai/evaluations/harvey-lab-aa",
      "exposure": "private",
      "unit": "percent",
      "sourceField": "harveyLab",
      "note": "Hallucination-gated all-pass rate. Do not mix v1.0 scores."
    },
    {
      "key": "itbench",
      "name": "ITBench-AA",
      "url": "https://artificialanalysis.ai/evaluations/itbench-aa",
      "exposure": "private",
      "unit": "percent",
      "sourceField": "itbenchSre"
    },
    {
      "key": "mlcr",
      "name": "MLCR-AA \u00b7 private holdout",
      "url": "https://artificialanalysis.ai/evaluations/mlcr-aa",
      "exposure": "private",
      "unit": "percent",
      "sourceField": "mlcrOverall",
      "note": "AA evaluates a private expert/compound held-out set, not the public MLCR dataset."
    },
    {
      "key": "scicode",
      "name": "SciCode",
      "url": "https://artificialanalysis.ai/evaluations/scicode",
      "exposure": "public",
      "unit": "percent",
      "sourceField": "scicode",
      "note": "Under review by Artificial Analysis after a dataset audit."
    },
    {
      "key": "terminalBench40",
      "name": "Terminal-Bench 4.0",
      "url": "https://artificialanalysis.ai/evaluations/terminalbench-4-0",
      "exposure": "public",
      "unit": "percent",
      "sourceField": "terminalBench40"
    },
    {
      "key": "terminalBench21",
      "name": "Terminal-Bench 2.1",
      "url": "https://artificialanalysis.ai/evaluations/terminalbench-2-1",
      "exposure": "public",
      "unit": "percent",
      "sourceField": "terminalBench21"
    },
    {
      "key": "terminalBenchScience",
      "name": "Terminal-Bench-Science 0.1",
      "url": "https://artificialanalysis.ai/evaluations/terminal-bench-science",
      "exposure": "public",
      "unit": "percent",
      "sourceField": "terminalBenchScience",
      "note": "Context only. Not used in the default pairings."
    },
    {
      "key": "gdpval",
      "name": "GDPval-AA v2.1",
      "url": "https://artificialanalysis.ai/evaluations/gdpval-aa",
      "exposure": "public",
      "unit": "elo",
      "sourceField": "gdpval",
      "note": "Raw Elo. Used as a predictor, never displayed as a percentage."
    },
    {
      "key": "gdpPdf",
      "name": "GDP.pdf \u00b7 all-pass",
      "url": "https://artificialanalysis.ai/evaluations/gdp-pdf",
      "exposure": "public",
      "unit": "percent",
      "sourceField": "gdpPdfAllPass"
    },
    {
      "key": "enterpriseOps",
      "name": "EnterpriseOps-Gym-AA",
      "url": "https://artificialanalysis.ai/evaluations/enterprise-ops-gym-aa",
      "exposure": "public",
      "unit": "percent",
      "sourceField": "enterpriseOpsGym"
    },
    {
      "key": "gpqa",
      "name": "GPQA Diamond",
      "url": "https://artificialanalysis.ai/evaluations/gpqa-diamond",
      "exposure": "public",
      "unit": "percent",
      "sourceField": "gpqa"
    },
    {
      "key": "hle",
      "name": "Humanity\u2019s Last Exam",
      "url": "https://artificialanalysis.ai/evaluations/humanitys-last-exam",
      "exposure": "public",
      "unit": "percent",
      "sourceField": "hle"
    },
    {
      "key": "lcr",
      "name": "AA-LCR v1.1",
      "url": "https://artificialanalysis.ai/evaluations/artificial-analysis-long-context-reasoning",
      "exposure": "public",
      "unit": "percent",
      "sourceField": "lcr"
    }
  ],
  "models": [
    {
      "id": "claude-opus-5-5",
      "name": "Claude Opus 5.5 (Max, Default Fallback)",
      "shortName": "Claude Opus 5.5 (max with fallback)",
      "provider": "Anthropic",
      "release": "2026-09-22",
      "scores": {
        "frontierCode": 52.17309999999999,
        "briefcaseRubric": 61.212121212121204,
        "automation": 69.5386812863607,
        "analyst": 56.25,
        "omniscienceAccuracy": 66.21666666666667,
        "harveyLab": 4.166666666666666,
        "itbench": 38.229755178907695,
        "mlcr": 66.6666666666667,
        "scicode": 66.8981481481482,
        "terminalBench40": 59.5959595959596,
        "terminalBench21": null,
        "terminalBenchScience": 59.047619047619,
        "gdpval": 1866.27,
        "gdpPdf": 26.200000000000003,
        "enterpriseOps": null,
        "gpqa": null,
        "hle": 61.353104726598694,
        "lcr": 84.6666666666667
      }
    },
    {
      "id": "claude-sonnet-5-5",
      "name": "Claude Sonnet 5.5 (Max, Default Fallback)",
      "shortName": "Claude Sonnet 5.5 (max with fallback)",
      "provider": "Anthropic",
      "release": "2026-09-28",
      "scores": {
        "frontierCode": 45.8291333333333,
        "briefcaseRubric": 65.15151515151516,
        "automation": 71.76882128193995,
        "analyst": 57.49999999999999,
        "omniscienceAccuracy": 53.949999999999996,
        "harveyLab": 2.7777777777777777,
        "itbench": null,
        "mlcr": 75.0,
        "scicode": 60.9953703703704,
        "terminalBench40": 63.636363636363605,
        "terminalBench21": null,
        "terminalBenchScience": 53.3333333333333,
        "gdpval": 1837.82,
        "gdpPdf": 25.8,
        "enterpriseOps": null,
        "gpqa": null,
        "hle": 54.9582947173309,
        "lcr": 82.6666666666667
      }
    },
    {
      "id": "claude-fable-5-1",
      "name": "Claude Fable 5.1 (Max, Default Fallback)",
      "shortName": "Claude Fable 5.1 (max with fallback)",
      "provider": "Anthropic",
      "release": "2026-09-01",
      "scores": {
        "frontierCode": 48.9732333333333,
        "briefcaseRubric": 61.51515151515151,
        "automation": 59.37591715646424,
        "analyst": 57.49999999999999,
        "omniscienceAccuracy": 67.23333333333333,
        "harveyLab": 6.388888888888888,
        "itbench": 49.5291902071563,
        "mlcr": 71.1111111111111,
        "scicode": 63.078703703703695,
        "terminalBench40": 52.020202020202,
        "terminalBench21": 91.3857677902622,
        "terminalBenchScience": 43.3333333333333,
        "gdpval": 1756.22,
        "gdpPdf": 26.200000000000003,
        "enterpriseOps": null,
        "gpqa": 93.7373737373737,
        "hle": 59.1288229842447,
        "lcr": 85.3333333333333
      }
    },
    {
      "id": "gpt-6-astra",
      "name": "GPT-6 Astra (Max)",
      "shortName": "GPT-6 Astra (max)",
      "provider": "OpenAI",
      "release": "2026-09-03",
      "scores": {
        "frontierCode": 46.9827666666667,
        "briefcaseRubric": 52.02020202020202,
        "automation": 68.49174806688876,
        "analyst": 51.24999999999999,
        "omniscienceAccuracy": 62.6,
        "harveyLab": 8.611111111111109,
        "itbench": 48.5875706214689,
        "mlcr": 35.0,
        "scicode": 56.4814814814815,
        "terminalBench40": 59.09090909090911,
        "terminalBench21": 88.38951310861421,
        "terminalBenchScience": 63.3333333333333,
        "gdpval": 1571.66,
        "gdpPdf": 31.0,
        "enterpriseOps": null,
        "gpqa": 96.0606060606061,
        "hle": 54.6802594995366,
        "lcr": 80.6666666666667
      }
    },
    {
      "id": "gpt-6-1-sol",
      "name": "GPT-6.1 Sol (Max)",
      "shortName": "GPT-6.1 Sol (max)",
      "provider": "OpenAI",
      "release": "2026-09-29",
      "scores": {
        "frontierCode": 49.1209333333333,
        "briefcaseRubric": 55.65656565656566,
        "automation": 64.86844882398171,
        "analyst": 50.0,
        "omniscienceAccuracy": 62.1,
        "harveyLab": 6.944444444444444,
        "itbench": null,
        "mlcr": 33.8888888888889,
        "scicode": 54.1666666666667,
        "terminalBench40": 56.0606060606061,
        "terminalBench21": null,
        "terminalBenchScience": 58.0952380952381,
        "gdpval": 1593.06,
        "gdpPdf": 31.0,
        "enterpriseOps": null,
        "gpqa": null,
        "hle": 52.919369786839695,
        "lcr": 83.0
      }
    },
    {
      "id": "muse-spark-1-3",
      "name": "Muse Spark 1.3 (Max)",
      "shortName": "Muse Spark 1.3 (max)",
      "provider": "Meta",
      "release": "2026-09-02",
      "scores": {
        "frontierCode": 40.6458333333333,
        "briefcaseRubric": 58.686868686868685,
        "automation": 57.853437865386134,
        "analyst": 37.5,
        "omniscienceAccuracy": 43.583333333333336,
        "harveyLab": 8.888888888888888,
        "itbench": 33.239171374764595,
        "mlcr": 43.3333333333333,
        "scicode": 58.796296296296305,
        "terminalBench40": 33.3333333333333,
        "terminalBench21": 84.2696629213483,
        "terminalBenchScience": 10.952380952380999,
        "gdpval": 1680.99,
        "gdpPdf": 26.6,
        "enterpriseOps": null,
        "gpqa": 93.53535353535351,
        "hle": 48.7025023169602,
        "lcr": 83.0
      }
    },
    {
      "id": "grok-4-7",
      "name": "Grok 4.7 (Xhigh)",
      "shortName": "Grok 4.7 (xhigh)",
      "provider": "SpaceXAI",
      "release": "2026-09-21",
      "scores": {
        "frontierCode": 33.834199999999996,
        "briefcaseRubric": 55.35353535353536,
        "automation": 65.55835148969457,
        "analyst": null,
        "omniscienceAccuracy": 47.449999999999996,
        "harveyLab": 9.444444444444446,
        "itbench": 42.0809792843691,
        "mlcr": 15.0,
        "scicode": 57.4074074074074,
        "terminalBench40": 25.7575757575758,
        "terminalBench21": null,
        "terminalBenchScience": null,
        "gdpval": 1712.45,
        "gdpPdf": 20.0,
        "enterpriseOps": null,
        "gpqa": null,
        "hle": 43.1417979610751,
        "lcr": 76.6666666666667
      }
    },
    {
      "id": "mimo-v2-6-pro",
      "name": "MiMo-V2.6-Pro",
      "shortName": "MiMo-V2.6-Pro",
      "provider": "Xiaomi",
      "release": "2026-09-21",
      "scores": {
        "frontierCode": 45.4153333333333,
        "briefcaseRubric": 54.74747474747475,
        "automation": 58.623409434190854,
        "analyst": 37.5,
        "omniscienceAccuracy": 34.849999999999994,
        "harveyLab": null,
        "itbench": null,
        "mlcr": 18.3333333333333,
        "scicode": 60.8796296296296,
        "terminalBench40": 34.8484848484849,
        "terminalBench21": null,
        "terminalBenchScience": 5.71428571428571,
        "gdpval": 1682.08,
        "gdpPdf": 19.2,
        "enterpriseOps": null,
        "gpqa": null,
        "hle": 49.351251158480096,
        "lcr": 86.3333333333333
      }
    },
    {
      "id": "qwen3-8-max",
      "name": "Qwen3.8 Max (0902)",
      "shortName": "Qwen3.8 Max (0902)",
      "provider": "Alibaba",
      "release": "2026-09-02",
      "scores": {
        "frontierCode": 43.156299999999995,
        "briefcaseRubric": 57.95574288724974,
        "automation": 56.1905381087678,
        "analyst": 45.0,
        "omniscienceAccuracy": 31.683333333333337,
        "harveyLab": null,
        "itbench": 40.254237288135606,
        "mlcr": 20.0,
        "scicode": 52.0833333333333,
        "terminalBench40": 38.8888888888889,
        "terminalBench21": 88.7640449438202,
        "terminalBenchScience": 11.9047619047619,
        "gdpval": 1667.28,
        "gdpPdf": 22.8,
        "enterpriseOps": 47.62757385854969,
        "gpqa": 92.82828282828281,
        "hle": 43.0954587581094,
        "lcr": 80.3333333333333
      }
    },
    {
      "id": "glm-5-3",
      "name": "GLM-5.3 (Max)",
      "shortName": "GLM-5.3 (max)",
      "provider": "Z AI",
      "release": "2026-08-18",
      "scores": {
        "frontierCode": 40.0752,
        "briefcaseRubric": 51.34854771784232,
        "automation": 62.202864996264196,
        "analyst": null,
        "omniscienceAccuracy": 33.85,
        "harveyLab": 0.27777777777777773,
        "itbench": 46.0828625235405,
        "mlcr": 48.3333333333333,
        "scicode": 59.0277777777778,
        "terminalBench40": 41.9191919191919,
        "terminalBench21": 83.8951310861423,
        "terminalBenchScience": 9.52380952380952,
        "gdpval": 1650.73,
        "gdpPdf": 11.200000000000001,
        "enterpriseOps": 36.436884512085946,
        "gpqa": 91.7171717171717,
        "hle": 42.2613531047266,
        "lcr": 79.6666666666667
      }
    },
    {
      "id": "step-5",
      "name": "Step 5 Preview",
      "shortName": "Step 5 Preview",
      "provider": "StepFun",
      "release": "2026-09-18",
      "scores": {
        "frontierCode": 38.4270666666667,
        "briefcaseRubric": 47.77777777777778,
        "automation": 50.956594190962875,
        "analyst": 35.0,
        "omniscienceAccuracy": 41.53333333333333,
        "harveyLab": null,
        "itbench": 55.5555555555556,
        "mlcr": 16.6666666666667,
        "scicode": 58.912037037037,
        "terminalBench40": 33.3333333333333,
        "terminalBench21": null,
        "terminalBenchScience": 2.38095238095238,
        "gdpval": 1584.24,
        "gdpPdf": 14.799999999999999,
        "enterpriseOps": 47.17994628469114,
        "gpqa": null,
        "hle": 46.4782205746061,
        "lcr": 88.3333333333333
      }
    },
    {
      "id": "kimi-k3",
      "name": "Kimi K3 (Max)",
      "shortName": "Kimi K3 (max)",
      "provider": "Kimi",
      "release": "2026-07-16",
      "scores": {
        "frontierCode": 43.6099333333333,
        "briefcaseRubric": 50.97236438075742,
        "automation": 58.272977800152034,
        "analyst": 38.75,
        "omniscienceAccuracy": 47.583333333333336,
        "harveyLab": 5.277777777777778,
        "itbench": 47.6930320150659,
        "mlcr": 38.3333333333333,
        "scicode": 59.4907407407407,
        "terminalBench40": 12.6262626262626,
        "terminalBench21": 85.0187265917603,
        "terminalBenchScience": null,
        "gdpval": 1535.63,
        "gdpPdf": 22.0,
        "enterpriseOps": 45.329752312742464,
        "gpqa": 93.53535353535351,
        "hle": 46.8952734012975,
        "lcr": 88.6666666666667
      }
    },
    {
      "id": "claude-haiku-5-5",
      "name": "Claude Haiku 5.5 (Max)",
      "shortName": "Claude Haiku 5.5 (max)",
      "provider": "Anthropic",
      "release": "2026-10-07",
      "scores": {
        "frontierCode": 41.8067,
        "briefcaseRubric": 52.323232323232325,
        "automation": 35.411015132783625,
        "analyst": 53.75,
        "omniscienceAccuracy": 36.38333333333333,
        "harveyLab": 0.5555555555555555,
        "itbench": null,
        "mlcr": 63.3333333333333,
        "scicode": 54.976851851851805,
        "terminalBench40": 32.8282828282828,
        "terminalBench21": null,
        "terminalBenchScience": 20.0,
        "gdpval": 1620.01,
        "gdpPdf": 20.8,
        "enterpriseOps": null,
        "gpqa": null,
        "hle": 44.3929564411492,
        "lcr": 82.6666666666667
      }
    },
    {
      "id": "glm-5-3-flash",
      "name": "GLM-5.3-Flash",
      "shortName": "GLM-5.3-Flash",
      "provider": "Z AI",
      "release": "2026-08-26",
      "scores": {
        "frontierCode": 40.9957333333333,
        "briefcaseRubric": 46.35897435897436,
        "automation": 60.36862782167782,
        "analyst": null,
        "omniscienceAccuracy": 27.500000000000004,
        "harveyLab": 0.8333333333333334,
        "itbench": 51.24293785310739,
        "mlcr": 51.11111111111109,
        "scicode": 51.6203703703704,
        "terminalBench40": 32.8282828282828,
        "terminalBench21": 84.2696629213483,
        "terminalBenchScience": 4.76190476190476,
        "gdpval": 1644.13,
        "gdpPdf": 15.4,
        "enterpriseOps": 33.21396598030439,
        "gpqa": 91.2121212121212,
        "hle": 39.8517145505097,
        "lcr": 80.0
      }
    },
    {
      "id": "gemini-3-8-flash",
      "name": "Gemini 3.8 Flash (High)",
      "shortName": "Gemini 3.8 Flash (high)",
      "provider": "Google",
      "release": "2026-09-02",
      "scores": {
        "frontierCode": 36.3585,
        "briefcaseRubric": 42.12121212121212,
        "automation": 59.93009432118242,
        "analyst": null,
        "omniscienceAccuracy": 54.6,
        "harveyLab": 0,
        "itbench": 52.5423728813559,
        "mlcr": 21.6666666666667,
        "scicode": 56.5972222222222,
        "terminalBench40": 19.6969696969697,
        "terminalBench21": 87.6404494382023,
        "terminalBenchScience": 10.0,
        "gdpval": 1435.52,
        "gdpPdf": 21.0,
        "enterpriseOps": null,
        "gpqa": 95.2525252525253,
        "hle": 47.8220574606117,
        "lcr": 81.3333333333333
      }
    },
    {
      "id": "deepseek-v4-1-flash",
      "name": "DeepSeek V4.1 Flash (Max)",
      "shortName": "DeepSeek V4.1 Flash (max)",
      "provider": "DeepSeek",
      "release": "2026-09-10",
      "scores": {
        "frontierCode": 44.0335333333333,
        "briefcaseRubric": 45.194274028629856,
        "automation": 68.89097769674058,
        "analyst": null,
        "omniscienceAccuracy": 46.400000000000006,
        "harveyLab": 1.6666666666666667,
        "itbench": 46.9161958568738,
        "mlcr": 22.7777777777778,
        "scicode": 51.851851851851805,
        "terminalBench40": 26.7676767676768,
        "terminalBench21": null,
        "terminalBenchScience": 9.047619047619051,
        "gdpval": 1600,
        "gdpPdf": 12.8,
        "enterpriseOps": null,
        "gpqa": null,
        "hle": 39.2493049119555,
        "lcr": 84.0
      }
    },
    {
      "id": "gpt-6-luna",
      "name": "GPT-6 Luna (Max)",
      "shortName": "GPT-6 Luna (max)",
      "provider": "OpenAI",
      "release": "2026-09-22",
      "scores": {
        "frontierCode": 41.9269666666667,
        "briefcaseRubric": 41.21212121212121,
        "automation": 53.16814034483175,
        "analyst": 43.75,
        "omniscienceAccuracy": 43.78333333333334,
        "harveyLab": 3.3333333333333335,
        "itbench": null,
        "mlcr": 16.1111111111111,
        "scicode": 54.6296296296296,
        "terminalBench40": 12.6262626262626,
        "terminalBench21": null,
        "terminalBenchScience": 8.57142857142857,
        "gdpval": 1430.2,
        "gdpPdf": 22.8,
        "enterpriseOps": null,
        "gpqa": null,
        "hle": 38.5078776645042,
        "lcr": 83.3333333333333
      }
    },
    {
      "id": "qwen3-8-27b",
      "name": "Qwen3.8 27B (Xhigh)",
      "shortName": "Qwen3.8 27B (xhigh)",
      "provider": "Alibaba",
      "release": "2026-08-14",
      "scores": {
        "frontierCode": 34.2324,
        "briefcaseRubric": 47.80388151174668,
        "automation": 48.24244368224193,
        "analyst": 18.75,
        "omniscienceAccuracy": 15.583333333333332,
        "harveyLab": null,
        "itbench": null,
        "mlcr": 21.6666666666667,
        "scicode": 46.6435185185185,
        "terminalBench40": 5.55555555555556,
        "terminalBench21": 79.7752808988764,
        "terminalBenchScience": 0.476190476190476,
        "gdpval": 1419.13,
        "gdpPdf": 16.6,
        "enterpriseOps": 44.225604297224706,
        "gpqa": 90.5050505050505,
        "hle": 33.920296570899,
        "lcr": 82.0
      }
    },
    {
      "id": "k2-horizon-375b-a23b",
      "name": "K2 Horizon 375B A23B",
      "shortName": "K2 Horizon 375B A23B",
      "provider": "Institute of Foundation Models",
      "release": "2026-09-03",
      "scores": {
        "frontierCode": 32.0955,
        "briefcaseRubric": 41.64926931106472,
        "automation": 37.20802754388014,
        "analyst": null,
        "omniscienceAccuracy": 18.183333333333334,
        "harveyLab": null,
        "itbench": null,
        "mlcr": null,
        "scicode": 42.939814814814795,
        "terminalBench40": 1.51515151515152,
        "terminalBench21": 71.91011235955061,
        "terminalBenchScience": null,
        "gdpval": 1360.51,
        "gdpPdf": 7.3999999999999995,
        "enterpriseOps": null,
        "gpqa": 87.27272727272731,
        "hle": 31.974050046339197,
        "lcr": 80.0
      }
    },
    {
      "id": "minimax-m3",
      "name": "MiniMax-M3",
      "shortName": "MiniMax-M3",
      "provider": "MiniMax",
      "release": "2026-06-01",
      "scores": {
        "frontierCode": 23.0037666666667,
        "briefcaseRubric": 30.303030303030305,
        "automation": 21.25125760074941,
        "analyst": 10.0,
        "omniscienceAccuracy": 16.7,
        "harveyLab": 0,
        "itbench": null,
        "mlcr": 17.2222222222222,
        "scicode": 47.1064814814815,
        "terminalBench40": 2.02020202020202,
        "terminalBench21": 65.1685393258427,
        "terminalBenchScience": 0.476190476190476,
        "gdpval": 1242.26,
        "gdpPdf": 9.8,
        "enterpriseOps": 32.10981796478663,
        "gpqa": 92.9292929292929,
        "hle": 38.9712696941613,
        "lcr": 83.0
      }
    },
    {
      "id": "inkling",
      "name": "Inkling (Xhigh)",
      "shortName": "Inkling (xhigh)",
      "provider": "Thinking Machines",
      "release": "2026-07-15",
      "scores": {
        "frontierCode": 15.8388333333333,
        "briefcaseRubric": 19.29292929292929,
        "automation": 4.984889684475201,
        "analyst": 23.75,
        "omniscienceAccuracy": 41.55,
        "harveyLab": 0,
        "itbench": null,
        "mlcr": 12.2222222222222,
        "scicode": 46.9907407407407,
        "terminalBench40": 1.01010101010101,
        "terminalBench21": 55.0561797752809,
        "terminalBenchScience": 0,
        "gdpval": 1075.6,
        "gdpPdf": 12.8,
        "enterpriseOps": 37.95881826320501,
        "gpqa": 87.1717171717172,
        "hle": 31.881371640407803,
        "lcr": 77.3333333333333
      }
    },
    {
      "id": "muse-glimmer",
      "name": "Muse Glimmer (High)",
      "shortName": "Muse Glimmer (high)",
      "provider": "Meta",
      "release": "2026-08-10",
      "scores": {
        "frontierCode": 13.917833333333299,
        "briefcaseRubric": 11.514522821576763,
        "automation": 6.831776269647931,
        "analyst": 20.0,
        "omniscienceAccuracy": 26.98333333333333,
        "harveyLab": null,
        "itbench": null,
        "mlcr": 20.0,
        "scicode": 44.907407407407405,
        "terminalBench40": 0.505050505050505,
        "terminalBench21": 51.685393258427005,
        "terminalBenchScience": 0,
        "gdpval": 786.36,
        "gdpPdf": 10.0,
        "enterpriseOps": 34.73589973142345,
        "gpqa": 83.5353535353535,
        "hle": 21.9647822057461,
        "lcr": 83.3333333333333
      }
    }
  ],
  "comparisons": [
    {
      "key": "frontierCode",
      "label": "Coding",
      "predictors": [
        "scicode",
        "terminalBench40"
      ]
    },
    {
      "key": "briefcaseRubric",
      "label": "Knowledge work",
      "predictors": [
        "gdpval",
        "gdpPdf"
      ]
    },
    {
      "key": "automation",
      "label": "Workflows",
      "predictors": [
        "enterpriseOps",
        "terminalBench40"
      ]
    },
    {
      "key": "analyst",
      "label": "Data analysis",
      "predictors": [
        "scicode",
        "gdpPdf"
      ]
    },
    {
      "key": "omniscienceAccuracy",
      "label": "Knowledge",
      "predictors": [
        "gpqa",
        "hle"
      ]
    },
    {
      "key": "harveyLab",
      "label": "Legal work",
      "predictors": [
        "gdpPdf",
        "lcr"
      ]
    },
    {
      "key": "itbench",
      "label": "Operations",
      "predictors": [
        "terminalBench40",
        "terminalBench21"
      ]
    },
    {
      "key": "mlcr",
      "label": "Medical reasoning",
      "predictors": [
        "lcr",
        "gdpPdf"
      ]
    }
  ],
  "cohort": "All 22 exact variants in the retrieved FrontierCode cohort. Other private evaluations are compared within this cohort. This is not all Artificial Analysis models.",
  "sources": [
    {
      "url": "https://artificialanalysis.ai/evaluations/frontiercode",
      "sha256": "791988e690cd0e94596067bc213b4a039fb035315a5e9dd340ac15cba2bfb146",
      "role": "All published score fields for the fixed cohort"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/aa-analyst-agent",
      "sha256": "741fbf1c59bafc034e124108e4866bf5b63707b06ba5dbc732be2df667168f3e",
      "role": "Dataset visibility and score definition"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/aa-briefcase",
      "sha256": "89b6f29835bfe650ff2f75bdf7a10ff73e4b0fd3f74f74c99dda85b39b04a53b",
      "role": "Dataset visibility and score definition"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/artificial-analysis-long-context-reasoning",
      "sha256": "de30eb88323a1ffeed76e74d867d1bf5e3452b6b9059e21c195b3e9769bf414b",
      "role": "Dataset visibility and score definition"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/automationbench-aa",
      "sha256": "2b18a8f504bff85180547fb9042ff6ee4ad6267aeb9ec64d2c849a6dbad6fc1d",
      "role": "Dataset visibility and score definition"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/enterprise-ops-gym-aa",
      "sha256": "aa673a34a071aa282126fc6f401c2414a433685a8d567bddd0443cdf173d4973",
      "role": "Dataset visibility and score definition"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/gdp-pdf",
      "sha256": "a4a17f6b03aedeab3bd311d9f45ddd2e1fefb7c26597a2160ed1bada50f02872",
      "role": "Dataset visibility and score definition"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/gdpval-aa",
      "sha256": "fd6bca5a9debcc27fcc0f5791473868f223c92c3f7af9305940f1a0fe738c9f5",
      "role": "Dataset visibility and score definition"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/gpqa-diamond",
      "sha256": "93009a70cfb65d09727008a6f2f906ac673608aabf6db7d87f07497cd8b22129",
      "role": "Dataset visibility and score definition"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/harvey-lab-aa",
      "sha256": "11841162a15e888a9b7a992301470d9ea2419f4e0688327aa4bb40b76f937977",
      "role": "Dataset visibility and score definition"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/humanitys-last-exam",
      "sha256": "2947cbadebfbe2f9c589ca52e4444c3495825ed71076e319713e61d271840562",
      "role": "Dataset visibility and score definition"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/itbench-aa",
      "sha256": "749e13adf3e851045fcd1a6d3c0d6d6c819d60d5798314dd20977933355ef771",
      "role": "Dataset visibility and score definition"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/mlcr-aa",
      "sha256": "4396bc1690eedd6ffb06b094769b255b60c75bf2bbf9ce1d5a18df4ae0c4fe10",
      "role": "Dataset visibility and score definition"
    },
    {
      "url": "https://artificialanalysis.ai/evaluations/omniscience",
      "sha256": "280764aace0ba446442909ad4bf7431c44ba70178c5abb0866432663a63d248b",
      "role": "Dataset visibility and score definition"
    }
  ]
}