{
  "generated_at": "2026-09-30T05:53:10.306Z",
  "output_ratio": 0.35,
  "sources": [
    "https://epoch.ai/benchmarks"
  ],
  "entries": [
    {
      "model_id": null,
      "label": "Amazon Nova Pro",
      "organization": "Amazon",
      "released": "2024-12-03",
      "scores": {
        "eci": 123.75,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-fable-5",
      "label": "Anthropic: Claude Fable 5",
      "organization": "Anthropic",
      "released": "2026-06-09",
      "scores": {
        "eci": 163.6,
        "deepswe": 69.91150442477876,
        "scicode": 60.9953703703704,
        "aime": 100,
        "gpqa": 85.85858585858585,
        "arc_agi_2": 89.16666666666667
      },
      "deepswe_task_usd": 13.414521495535714,
      "cost_per_1m_usd": 24,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-fable-5.1",
      "label": "Anthropic: Claude Fable 5.1",
      "organization": "Anthropic",
      "released": "2026-09-01",
      "scores": {
        "eci": 165,
        "deepswe": null,
        "scicode": 63.078703703703695,
        "aime": 100,
        "gpqa": null,
        "arc_agi_2": 90
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 24,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-haiku-4.5",
      "label": "Anthropic: Claude Haiku 4.5",
      "organization": "Anthropic",
      "released": "2025-10-15",
      "scores": {
        "eci": 142.42,
        "deepswe": null,
        "scicode": 43.287037037037,
        "aime": 66.66666666666666,
        "gpqa": 71.21212121212122,
        "arc_agi_2": 4.03
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 2.4,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-opus-4.1",
      "label": "Anthropic: Claude Opus 4.1",
      "organization": "Anthropic",
      "released": "2025-08-05",
      "scores": {
        "eci": 144.11,
        "deepswe": null,
        "scicode": null,
        "aime": 68.88888888888889,
        "gpqa": 77.27272727272727,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 36,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-opus-4.5",
      "label": "Anthropic: Claude Opus 4.5",
      "organization": "Anthropic",
      "released": "2025-11-24",
      "scores": {
        "eci": 150.1,
        "deepswe": null,
        "scicode": null,
        "aime": 86.11111111111111,
        "gpqa": 86.0479797979798,
        "arc_agi_2": 37.64
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 12,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-opus-4.6",
      "label": "Anthropic: Claude Opus 4.6",
      "organization": "Anthropic",
      "released": "2026-02-05",
      "scores": {
        "eci": 155.36,
        "deepswe": null,
        "scicode": null,
        "aime": 94.44444444444444,
        "gpqa": 90.53030303030303,
        "arc_agi_2": 69.17
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 12,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-opus-4.7",
      "label": "Anthropic: Claude Opus 4.7",
      "organization": "Anthropic",
      "released": "2026-04-16",
      "scores": {
        "eci": 156.39,
        "deepswe": null,
        "scicode": 54.51388888888889,
        "aime": 97.8,
        "gpqa": 90.15151515151516,
        "arc_agi_2": 75.83
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 12,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-opus-4.8",
      "label": "Anthropic: Claude Opus 4.8",
      "organization": "Anthropic",
      "released": "2026-05-28",
      "scores": {
        "eci": 158.3,
        "deepswe": 58.97435897435898,
        "scicode": 53.4722222222222,
        "aime": 98.33333333333333,
        "gpqa": 91.03535353535354,
        "arc_agi_2": 72.08
      },
      "deepswe_task_usd": 13.222583593240094,
      "cost_per_1m_usd": 12,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-opus-5",
      "label": "Anthropic: Claude Opus 5",
      "organization": "Anthropic",
      "released": "2026-07-24",
      "scores": {
        "eci": 162.67,
        "deepswe": 73.64864864864865,
        "scicode": 56.3657407407407,
        "aime": 98.88888888888889,
        "gpqa": 93.87626262626263,
        "arc_agi_2": 90.42
      },
      "deepswe_task_usd": 11.837583271396396,
      "cost_per_1m_usd": 12,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-opus-5.5",
      "label": "Anthropic: Claude Opus 5.5",
      "organization": "Anthropic",
      "released": "2026-09-22",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 66.8981481481482,
        "aime": 100,
        "gpqa": 90.59343434343434,
        "arc_agi_2": 92.5
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 9.6,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-sonnet-4",
      "label": "Anthropic: Claude Sonnet 4",
      "organization": "Anthropic",
      "released": "2025-05-22",
      "scores": {
        "eci": 141.69,
        "deepswe": null,
        "scicode": 40.046296296296305,
        "aime": 71.11111111111111,
        "gpqa": 79.18781725888326,
        "arc_agi_2": 5.93
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 7.2,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-sonnet-4.5",
      "label": "Anthropic: Claude Sonnet 4.5",
      "organization": "Anthropic",
      "released": "2025-09-29",
      "scores": {
        "eci": 146.84,
        "deepswe": null,
        "scicode": 44.6759259259259,
        "aime": 77.77777777777779,
        "gpqa": 82.32323232323232,
        "arc_agi_2": 13.61
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 7.2,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-sonnet-4.6",
      "label": "Anthropic: Claude Sonnet 4.6",
      "organization": "Anthropic",
      "released": "2026-02-17",
      "scores": {
        "eci": 152.25,
        "deepswe": 29.933481152993345,
        "scicode": 46.7592592592593,
        "aime": 85.7936507936508,
        "gpqa": 87.37373737373737,
        "arc_agi_2": 60.419999999999995
      },
      "deepswe_task_usd": 5.5223517987804875,
      "cost_per_1m_usd": 7.2,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-sonnet-5",
      "label": "Anthropic: Claude Sonnet 5",
      "organization": "Anthropic",
      "released": "2026-06-30",
      "scores": {
        "eci": 156.34,
        "deepswe": 53.84615384615385,
        "scicode": 53.587962962963,
        "aime": 94.72222222222221,
        "gpqa": 90.53030303030303,
        "arc_agi_2": null
      },
      "deepswe_task_usd": 26.399858950791852,
      "cost_per_1m_usd": 4.8,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "anthropic/claude-sonnet-5.5",
      "label": "Anthropic: Claude Sonnet 5.5",
      "organization": "Anthropic",
      "released": "2026-09-28",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 60.9953703703704,
        "aime": 100,
        "gpqa": 95.58080808080808,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 4.8,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "arcee-ai/trinity-large-thinking",
      "label": "Arcee AI: Trinity Large Thinking",
      "organization": "Arcee AI",
      "released": "2026-04-01",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 36.1111111111111,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.4425,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Baichuan 2-7B",
      "organization": "Baichuan",
      "released": "2023-09-20",
      "scores": {
        "eci": 95.75,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Baichuan1-7B",
      "organization": "Baichuan",
      "released": "2023-06-01",
      "scores": {
        "eci": 89.77,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Baichuan2-13B",
      "organization": "Baichuan",
      "released": "2023-09-06",
      "scores": {
        "eci": 102.76,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Cerebras-GPT-13B",
      "organization": "Cerebras Systems",
      "released": "2023-03-20",
      "scores": {
        "eci": 82.36,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "chatglm2-6b",
      "organization": null,
      "released": "2023-06-24",
      "scores": {
        "eci": 98.41,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Claude 2",
      "organization": "Anthropic",
      "released": "2023-07-11",
      "scores": {
        "eci": 120.08,
        "deepswe": null,
        "scicode": null,
        "aime": 2.5,
        "gpqa": 34.659090909090914,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Claude 2.1",
      "organization": "Anthropic",
      "released": "2023-11-21",
      "scores": {
        "eci": 119.22,
        "deepswe": null,
        "scicode": null,
        "aime": 1.9444444444444444,
        "gpqa": 32.95454545454545,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Claude 3 Haiku",
      "organization": "Anthropic",
      "released": "2024-03-07",
      "scores": {
        "eci": 118.3,
        "deepswe": null,
        "scicode": null,
        "aime": 1.8055555555555554,
        "gpqa": 36.30050505050505,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Claude 3 Opus",
      "organization": "Anthropic",
      "released": "2024-02-29",
      "scores": {
        "eci": 126.91,
        "deepswe": null,
        "scicode": null,
        "aime": 4.722222222222222,
        "gpqa": 47.159090909090914,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Claude 3 Sonnet",
      "organization": "Anthropic",
      "released": "2024-02-29",
      "scores": {
        "eci": 120.69,
        "deepswe": null,
        "scicode": null,
        "aime": 2.5,
        "gpqa": 40.59343434343434,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Claude 3.5 Haiku",
      "organization": "Anthropic",
      "released": "2024-10-22",
      "scores": {
        "eci": 127.15,
        "deepswe": null,
        "scicode": 27.430555555555603,
        "aime": 4.305555555555555,
        "gpqa": 38.131313131313135,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Claude 3.5 Sonnet (October 2024)",
      "organization": "Anthropic",
      "released": "2024-06-20",
      "scores": {
        "eci": 130,
        "deepswe": null,
        "scicode": null,
        "aime": 6.527777777777779,
        "gpqa": 54.04040404040404,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Claude 3.7 Sonnet",
      "organization": "Anthropic",
      "released": "2025-02-24",
      "scores": {
        "eci": 141.16,
        "deepswe": null,
        "scicode": null,
        "aime": 57.77777777777777,
        "gpqa": 79.73484848484848,
        "arc_agi_2": 0.8999999999999999
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Claude Instant",
      "organization": "Anthropic",
      "released": "2023-08-09",
      "scores": {
        "eci": 120.2,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Claude Opus 4",
      "organization": "Anthropic",
      "released": "2025-05-22",
      "scores": {
        "eci": 142.67,
        "deepswe": null,
        "scicode": null,
        "aime": 64.44444444444444,
        "gpqa": 76.26262626262627,
        "arc_agi_2": 8.61
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "CodeQwen1.5-7B",
      "organization": null,
      "released": "2024-04-15",
      "scores": {
        "eci": 94.24,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "codex-mini-2025-05-16",
      "organization": null,
      "released": "2025-05-16",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": 1.27
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Cogito v2.1",
      "organization": "Deep Cogito",
      "released": "2025-11-19",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 40.9722222222222,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "cohere/command-a-plus",
      "label": "Cohere: Command A+",
      "organization": "Cohere,Cohere Labs (formerly Cohere for AI)",
      "released": "2026-05-20",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 37.8472222222222,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.72,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "cohere/command-r-08-2024",
      "label": "Cohere: Command R (08-2024)",
      "organization": "Cohere,Cohere Labs (formerly Cohere for AI)",
      "released": "2024-08-30",
      "scores": {
        "eci": 119.28,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.3075,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "DBRX",
      "organization": "Databricks",
      "released": "2024-03-27",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 32.891414141414145,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "DeepSeek Coder 1.3B",
      "organization": "DeepSeek,Peking University",
      "released": "2023-11-02",
      "scores": {
        "eci": 62.18,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "DeepSeek Coder 33B",
      "organization": "DeepSeek,Peking University",
      "released": "2023-11-02",
      "scores": {
        "eci": 95.74,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "DeepSeek Coder 6.7B",
      "organization": "DeepSeek,Peking University",
      "released": "2023-11-02",
      "scores": {
        "eci": 88.88,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "DeepSeek LLM 67B",
      "organization": "DeepSeek",
      "released": "2023-11-29",
      "scores": {
        "eci": 110.54,
        "deepswe": null,
        "scicode": null,
        "aime": 0.8333333333333334,
        "gpqa": 24.62121212121212,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "DeepSeek-Coder-V2-Lite-Base",
      "organization": null,
      "released": "2024-06-13",
      "scores": {
        "eci": 108.57,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "deepseek-r1-0528-qwen3-8b",
      "organization": null,
      "released": "2025-05-29",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 43.888888888888886,
        "gpqa": 9.280303030303031,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "DeepSeek-R1-Distill-Llama-70B",
      "organization": "DeepSeek",
      "released": "2025-01-20",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 51.388888888888886,
        "gpqa": 55.744949494949495,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "DeepSeek-R1-Distill-Qwen-1.5B",
      "organization": "DeepSeek",
      "released": "2025-01-20",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 21.38888888888889,
        "gpqa": 33.58585858585859,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "DeepSeek-R1-Distill-Qwen-14B",
      "organization": "DeepSeek",
      "released": "2025-01-20",
      "scores": {
        "eci": 135.43,
        "deepswe": null,
        "scicode": null,
        "aime": 50.55555555555556,
        "gpqa": 44.696969696969695,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "DeepSeek-R1-Distill-Qwen-32B",
      "organization": "DeepSeek",
      "released": "2025-01-20",
      "scores": {
        "eci": 137.43,
        "deepswe": null,
        "scicode": null,
        "aime": 55.55555555555556,
        "gpqa": 64.14141414141415,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "~deepseek/deepseek-pro-latest",
      "label": "DeepSeek: DeepSeek Pro Latest",
      "organization": "DeepSeek",
      "released": "2026-04-24",
      "scores": {
        "eci": 149.14,
        "deepswe": null,
        "scicode": 50,
        "aime": 96.66666666666667,
        "gpqa": 90.9090909090909,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.355,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "deepseek/deepseek-chat",
      "label": "DeepSeek: DeepSeek V3",
      "organization": "DeepSeek",
      "released": "2024-05-07",
      "scores": {
        "eci": 124.78,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.527355,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "deepseek/deepseek-chat-v3.1",
      "label": "DeepSeek: DeepSeek V3.1",
      "organization": "DeepSeek",
      "released": "2025-08-21",
      "scores": {
        "eci": 139.92,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.495,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "deepseek/deepseek-v3.1-terminus",
      "label": "DeepSeek: DeepSeek V3.1 Terminus",
      "organization": "DeepSeek",
      "released": "2025-09-22",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 40.625,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.5449999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "deepseek/deepseek-v3.2-exp",
      "label": "DeepSeek: DeepSeek V3.2 Exp",
      "organization": "DeepSeek",
      "released": "2025-09-29",
      "scores": {
        "eci": 145.02,
        "deepswe": null,
        "scicode": 38.8888888888889,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.319,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "deepseek/deepseek-v4-flash-0731",
      "label": "DeepSeek: DeepSeek V4 Flash 0731",
      "organization": "DeepSeek",
      "released": "2026-07-31",
      "scores": {
        "eci": 154.51,
        "deepswe": null,
        "scicode": 49.88425925925931,
        "aime": 94.44444444444444,
        "gpqa": 91.03535353535354,
        "arc_agi_2": 61.38888888888888
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.45449999999999996,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "~deepseek/deepseek-v4-flash-latest",
      "label": "DeepSeek: DeepSeek V4 Flash Latest",
      "organization": "DeepSeek",
      "released": "2026-04-24",
      "scores": {
        "eci": 146.1,
        "deepswe": null,
        "scicode": 44.907407407407405,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.45449999999999996,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "deepseek/deepseek-v4-pro-0813",
      "label": "DeepSeek: DeepSeek V4 Pro 0813",
      "organization": "DeepSeek",
      "released": "2026-08-13",
      "scores": {
        "eci": 155.39,
        "deepswe": null,
        "scicode": 51.0416666666667,
        "aime": 98.61111111111111,
        "gpqa": 91.66666666666666,
        "arc_agi_2": 61.25000000000001
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.122,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "deepseek/deepseek-v4.1-flash",
      "label": "DeepSeek: DeepSeek V4.1 Flash",
      "organization": "DeepSeek",
      "released": "2026-09-09",
      "scores": {
        "eci": 155.01,
        "deepswe": null,
        "scicode": 51.851851851851805,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.615,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "deepseek/deepseek-r1",
      "label": "DeepSeek: R1",
      "organization": "DeepSeek",
      "released": "2025-01-20",
      "scores": {
        "eci": 138.98,
        "deepswe": null,
        "scicode": 35.7142857142857,
        "aime": 53.333333333333336,
        "gpqa": 71.71717171717171,
        "arc_agi_2": 1.3
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.33,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Devstral Small 2",
      "organization": "Mistral AI",
      "released": "2025-12-09",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 28.819444444444397,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Dolly 2.0-12b",
      "organization": "Databricks",
      "released": "2023-04-11",
      "scores": {
        "eci": 88.94,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Eurus-2-7B-PRIME",
      "organization": "Tsinghua University,University of Illinois Urbana-Champaign (UIUC),Shanghai AI Lab,Peking University,Shanghai Jiao Tong University,CUHK Shenzhen Research Institute",
      "released": "2024-12-31",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 33.90151515151515,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Falcon 2 11B",
      "organization": "Technology Innovation Institute",
      "released": "2024-05-09",
      "scores": {
        "eci": 109.25,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Falcon-180B",
      "organization": "Technology Innovation Institute",
      "released": "2023-09-06",
      "scores": {
        "eci": 111.89,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Falcon-40B",
      "organization": "Technology Innovation Institute",
      "released": "2023-05-25",
      "scores": {
        "eci": 104.04,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Falcon-7B",
      "organization": "Technology Innovation Institute",
      "released": "2023-04-24",
      "scores": {
        "eci": 94.49,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Gemini 1.0 Pro",
      "organization": "Google DeepMind",
      "released": "2023-12-13",
      "scores": {
        "eci": 116.97,
        "deepswe": null,
        "scicode": null,
        "aime": 1.1111111111111112,
        "gpqa": 33.964646464646464,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Gemini 1.5 Flash (Sep 2024)",
      "organization": "Google DeepMind",
      "released": "2024-05-23",
      "scores": {
        "eci": 122.57,
        "deepswe": null,
        "scicode": null,
        "aime": 3.888888888888889,
        "gpqa": 40.37247474747475,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Gemini 1.5 Flash 8B",
      "organization": "Google DeepMind",
      "released": "2024-10-03",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 4.583333333333333,
        "gpqa": 32.95454545454545,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Gemini 1.5 Pro (Sept 2024)",
      "organization": "Google DeepMind",
      "released": "2024-05-14",
      "scores": {
        "eci": 126.9,
        "deepswe": null,
        "scicode": null,
        "aime": 6.805555555555555,
        "gpqa": 45.86489898989899,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Gemini 2.0 Flash Thinking (Jan 2025)",
      "organization": "Google DeepMind,Google",
      "released": "2025-02-05",
      "scores": {
        "eci": 134.69,
        "deepswe": null,
        "scicode": null,
        "aime": 31.11111111111111,
        "gpqa": 64.14141414141415,
        "arc_agi_2": 1.3
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Gemini 2.0 Pro",
      "organization": "Google DeepMind",
      "released": "2025-02-05",
      "scores": {
        "eci": 135.06,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 65.65656565656566,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Gemini 3 Deep Think",
      "organization": "Google DeepMind",
      "released": "2026-02-12",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": 84.58
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Gemini 3 Pro",
      "organization": "Google DeepMind",
      "released": "2025-11-18",
      "scores": {
        "eci": 152.96,
        "deepswe": null,
        "scicode": null,
        "aime": 91.38888888888889,
        "gpqa": 92.61363636363636,
        "arc_agi_2": 31.11
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Gemma 2 9B",
      "organization": "Google DeepMind",
      "released": "2024-06-24",
      "scores": {
        "eci": 119.78,
        "deepswe": null,
        "scicode": null,
        "aime": 0.5555555555555556,
        "gpqa": 27.46212121212121,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Gemma 2B",
      "organization": "Google DeepMind",
      "released": "2024-02-21",
      "scores": {
        "eci": 93.57,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Gemma 3 1B",
      "organization": "Google DeepMind",
      "released": "2025-03-12",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 1.1111111111111112,
        "gpqa": 19.94949494949495,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Gemma 7B",
      "organization": "Google DeepMind",
      "released": "2024-02-21",
      "scores": {
        "eci": 111.74,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "glm-4.7-flash_none",
      "organization": null,
      "released": null,
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 25,
        "gpqa": 45.13888888888889,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemini-2.5-flash",
      "label": "Google: Gemini 2.5 Flash",
      "organization": "Google DeepMind",
      "released": "2025-04-17",
      "scores": {
        "eci": 139.96,
        "deepswe": null,
        "scicode": null,
        "aime": 73.05555555555556,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.07,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemini-2.5-flash-lite",
      "label": "Google: Gemini 2.5 Flash Lite",
      "organization": "Google DeepMind",
      "released": "2025-06-17",
      "scores": {
        "eci": 133.93,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.205,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemini-2.5-pro",
      "label": "Google: Gemini 2.5 Pro",
      "organization": "Google DeepMind",
      "released": "2025-05-06",
      "scores": {
        "eci": 142.46,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 66.66666666666666,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 4.3125,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemini-3-flash-preview",
      "label": "Google: Gemini 3 Flash Preview",
      "organization": "Google DeepMind",
      "released": "2025-12-17",
      "scores": {
        "eci": 151.83,
        "deepswe": null,
        "scicode": null,
        "aime": 95.55555555555556,
        "gpqa": 89.39393939393939,
        "arc_agi_2": 33.61
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.3749999999999998,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemini-3.1-flash-lite",
      "label": "Google: Gemini 3.1 Flash Lite",
      "organization": "Google",
      "released": "2026-03-03",
      "scores": {
        "eci": 144.45,
        "deepswe": null,
        "scicode": 41.8981481481481,
        "aime": 80,
        "gpqa": 81.81818181818183,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.6874999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemini-3.1-pro-preview",
      "label": "Google: Gemini 3.1 Pro Preview",
      "organization": "Google DeepMind",
      "released": "2026-02-19",
      "scores": {
        "eci": 154.92,
        "deepswe": 11.725663716814159,
        "scicode": 58.912037037037,
        "aime": 95.6,
        "gpqa": 94.44444444444444,
        "arc_agi_2": 77.10000000000001
      },
      "deepswe_task_usd": 2.1434045568888886,
      "cost_per_1m_usd": 5.499999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemini-3.5-flash",
      "label": "Google: Gemini 3.5 Flash",
      "organization": "Google DeepMind",
      "released": "2026-05-19",
      "scores": {
        "eci": 154.55,
        "deepswe": 37.389380530973455,
        "scicode": 53.125,
        "aime": 95.55555555555556,
        "gpqa": 92.8030303030303,
        "arc_agi_2": 72.08
      },
      "deepswe_task_usd": 7.341823985508849,
      "cost_per_1m_usd": 4.125,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemini-3.5-flash-lite",
      "label": "Google: Gemini 3.5 Flash Lite",
      "organization": "Google DeepMind",
      "released": "2026-07-21",
      "scores": {
        "eci": 145.12,
        "deepswe": null,
        "scicode": 41.3194444444444,
        "aime": 71.11111111111111,
        "gpqa": 83.33333333333334,
        "arc_agi_2": 10.277777777777779
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.07,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemini-3.6-flash",
      "label": "Google: Gemini 3.6 Flash",
      "organization": "Google DeepMind",
      "released": "2026-07-21",
      "scores": {
        "eci": 154.36,
        "deepswe": 46.68141592920354,
        "scicode": 52.662037037037,
        "aime": 94.16666666666667,
        "gpqa": 94.12878787878788,
        "arc_agi_2": 60.416666666666664
      },
      "deepswe_task_usd": 4.419013277333334,
      "cost_per_1m_usd": 1.8,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemini-3.7-flash",
      "label": "Google: Gemini 3.7 Flash",
      "organization": "Google DeepMind",
      "released": "2026-08-13",
      "scores": {
        "eci": 157.72,
        "deepswe": 65.48672566371681,
        "scicode": 59.837962962963,
        "aime": 97.22222222222221,
        "gpqa": 94.82323232323232,
        "arc_agi_2": 84.58333333333333
      },
      "deepswe_task_usd": 2.0251351826880533,
      "cost_per_1m_usd": 1.8,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemini-3.8-flash",
      "label": "Google: Gemini 3.8 Flash",
      "organization": "Google DeepMind",
      "released": "2026-09-02",
      "scores": {
        "eci": 157.13,
        "deepswe": 73.8255033557047,
        "scicode": 56.5972222222222,
        "aime": 98.88888888888889,
        "gpqa": 95.39141414141415,
        "arc_agi_2": null
      },
      "deepswe_task_usd": 2.362349413758389,
      "cost_per_1m_usd": 1.8,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemma-2-27b-it",
      "label": "Google: Gemma 2 27B",
      "organization": "Google DeepMind",
      "released": "2024-06-24",
      "scores": {
        "eci": 122.06,
        "deepswe": null,
        "scicode": null,
        "aime": 1.3888888888888888,
        "gpqa": 36.48989898989899,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.65,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemma-3-12b-it",
      "label": "Google: Gemma 3 12B",
      "organization": "Google DeepMind",
      "released": "2025-03-12",
      "scores": {
        "eci": 123.49,
        "deepswe": null,
        "scicode": 17.3611111111111,
        "aime": 16.666666666666664,
        "gpqa": 39.457070707070706,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.08499999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemma-3-27b-it",
      "label": "Google: Gemma 3 27B",
      "organization": "Google DeepMind",
      "released": "2025-03-12",
      "scores": {
        "eci": 130.04,
        "deepswe": null,
        "scicode": 21.1805555555556,
        "aime": 22.5,
        "gpqa": 47.72727272727273,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.20949999999999996,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemma-3-4b-it",
      "label": "Google: Gemma 3 4B",
      "organization": "Google DeepMind",
      "released": "2025-03-12",
      "scores": {
        "eci": 116.02,
        "deepswe": null,
        "scicode": null,
        "aime": 7.5,
        "gpqa": 23.232323232323232,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.0675,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemma-4-26b-a4b-it",
      "label": "Google: Gemma 4 26B A4B ",
      "organization": "Google DeepMind",
      "released": "2026-04-02",
      "scores": {
        "eci": 141.85,
        "deepswe": null,
        "scicode": 40.046296296296305,
        "aime": 82.22222222222221,
        "gpqa": 73.23232323232324,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.138975,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "google/gemma-4-31b-it",
      "label": "Google: Gemma 4 31B",
      "organization": "Google DeepMind",
      "released": "2026-04-02",
      "scores": {
        "eci": 142.74,
        "deepswe": null,
        "scicode": 43.4027777777778,
        "aime": 73.33333333333333,
        "gpqa": 75.75757575757575,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.1775,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "GPT-4.5",
      "organization": "OpenAI",
      "released": "2025-02-27",
      "scores": {
        "eci": 136.74,
        "deepswe": null,
        "scicode": null,
        "aime": 37.77777777777778,
        "gpqa": 68.68686868686868,
        "arc_agi_2": 0.8
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "GPT-5.5 Instant",
      "organization": "OpenAI",
      "released": "2026-05-05",
      "scores": {
        "eci": 142.48,
        "deepswe": null,
        "scicode": 48.6111111111111,
        "aime": 68.05555555555556,
        "gpqa": 82.51262626262627,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Granite 4.1 30B",
      "organization": "IBM",
      "released": "2026-04-29",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 25.8101851851852,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "granite-4.0-1b",
      "organization": null,
      "released": "2025-10-28",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 0.8333333333333334,
        "gpqa": 23.98989898989899,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "granite-4.0-350m",
      "organization": null,
      "released": "2025-10-28",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 0.2777777777777778,
        "gpqa": 11.237373737373737,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Grok 3",
      "organization": "xAI",
      "released": "2025-04-09",
      "scores": {
        "eci": 138.34,
        "deepswe": null,
        "scicode": null,
        "aime": 55.55555555555556,
        "gpqa": 75.75757575757575,
        "arc_agi_2": 0
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Grok 4",
      "organization": "xAI",
      "released": "2025-07-09",
      "scores": {
        "eci": 146.46,
        "deepswe": null,
        "scicode": null,
        "aime": 84,
        "gpqa": 87,
        "arc_agi_2": 15.975
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Grok 4 Fast",
      "organization": "xAI",
      "released": "2025-09-19",
      "scores": {
        "eci": 144.21,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": 5.28
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Grok 4.3 Beta",
      "organization": "xAI",
      "released": "2026-04-17",
      "scores": {
        "eci": 149.13,
        "deepswe": null,
        "scicode": 47.337962962963,
        "aime": 93.33333333333333,
        "gpqa": 88.82575757575758,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Grok-2 (Dec 2024)",
      "organization": "xAI",
      "released": "2024-12-12",
      "scores": {
        "eci": 130.48,
        "deepswe": null,
        "scicode": null,
        "aime": 11.527777777777779,
        "gpqa": 53.78787878787878,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Grok-3 mini",
      "organization": "xAI",
      "released": "2025-06-24",
      "scores": {
        "eci": 140.35,
        "deepswe": null,
        "scicode": null,
        "aime": 77.77777777777779,
        "gpqa": 76.26262626262627,
        "arc_agi_2": 0.42
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Hermes 2 Theta Llama-3 70B",
      "organization": "Nous Research,Arcee AI",
      "released": "2024-06-20",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 2.5,
        "gpqa": 37.46843434343434,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "ibm-granite/granite-4.0-h-micro",
      "label": "IBM: Granite 4.0 Micro",
      "organization": null,
      "released": "2025-10-02",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 2.7777777777777777,
        "gpqa": 28.28282828282828,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.05025,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "inception/mercury-2",
      "label": "Inception: Mercury 2",
      "organization": "Inception Labs",
      "released": "2026-02-20",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 38.657407407407405,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.42499999999999993,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "inception/mercury-2.5",
      "label": "Inception: Mercury 2.5",
      "organization": "Inception Labs",
      "released": "2026-09-08",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.0785,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "INTELLECT-1",
      "organization": "Prime Intellect,Hugging Face,Arcee AI",
      "released": "2024-11-29",
      "scores": {
        "eci": 100.38,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "internlm-20b",
      "organization": null,
      "released": "2023-09-18",
      "scores": {
        "eci": 111.84,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "internlm-7b",
      "organization": null,
      "released": "2023-07-05",
      "scores": {
        "eci": 102.43,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Llama 2-13B",
      "organization": "Meta AI",
      "released": "2023-07-18",
      "scores": {
        "eci": 105.8,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Llama 2-34B",
      "organization": "Meta AI",
      "released": "2023-07-18",
      "scores": {
        "eci": 104.81,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Llama 2-70B",
      "organization": "Meta AI",
      "released": "2023-07-18",
      "scores": {
        "eci": 113.59,
        "deepswe": null,
        "scicode": null,
        "aime": 0,
        "gpqa": 26.325757575757574,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Llama 2-7B",
      "organization": "Meta AI",
      "released": "2023-07-18",
      "scores": {
        "eci": 98.54,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Llama 3-70B",
      "organization": "Meta AI",
      "released": "2024-04-18",
      "scores": {
        "eci": 122.91,
        "deepswe": null,
        "scicode": null,
        "aime": 4.305555555555555,
        "gpqa": 40.561868686868685,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Llama 3-8B",
      "organization": "Meta AI",
      "released": "2024-04-18",
      "scores": {
        "eci": 116.33,
        "deepswe": null,
        "scicode": null,
        "aime": 1.9444444444444444,
        "gpqa": 26.073232323232325,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Llama 3.1-405B",
      "organization": "Meta AI",
      "released": "2024-07-23",
      "scores": {
        "eci": 128.76,
        "deepswe": null,
        "scicode": null,
        "aime": 9.722222222222223,
        "gpqa": 50.91540404040404,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Llama 3.2 90B",
      "organization": "Meta AI",
      "released": "2024-09-24",
      "scores": {
        "eci": 125.5,
        "deepswe": null,
        "scicode": null,
        "aime": 2.638888888888889,
        "gpqa": 41.035353535353536,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "LLaMA-13B",
      "organization": "Meta AI",
      "released": "2023-02-24",
      "scores": {
        "eci": 100.09,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "LLaMA-33B",
      "organization": "Meta AI",
      "released": "2023-02-24",
      "scores": {
        "eci": 107.06,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "LLaMA-65B",
      "organization": "Meta AI",
      "released": "2023-02-24",
      "scores": {
        "eci": 109.88,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "LLaMA-7B",
      "organization": "Meta AI",
      "released": "2023-02-24",
      "scores": {
        "eci": 96.12,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Magistral Medium 1.1",
      "organization": "Mistral AI",
      "released": "2025-06-10",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": 0
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Magistral Medium 1.2",
      "organization": "Mistral AI",
      "released": "2025-09-18",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 39.2361111111111,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Magistral Small 1.0",
      "organization": "Mistral AI",
      "released": "2025-06-10",
      "scores": {
        "eci": 133.19,
        "deepswe": null,
        "scicode": null,
        "aime": 30,
        "gpqa": 56.060606060606055,
        "arc_agi_2": 0
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Magistral Small 1.2",
      "organization": "Mistral AI",
      "released": "2025-09-18",
      "scores": {
        "eci": 131.41,
        "deepswe": null,
        "scicode": 35.185185185185205,
        "aime": 28.055555555555557,
        "gpqa": 47.6010101010101,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "meta-llama/llama-3.1-70b-instruct",
      "label": "Meta: Llama 3.1 70B Instruct",
      "organization": "Meta AI",
      "released": "2024-07-23",
      "scores": {
        "eci": 125.91,
        "deepswe": null,
        "scicode": null,
        "aime": 3.6111111111111107,
        "gpqa": 44.19191919191919,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.4,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "meta-llama/llama-3.1-8b-instruct",
      "label": "Meta: Llama 3.1 8B Instruct",
      "organization": "Meta AI",
      "released": "2024-07-23",
      "scores": {
        "eci": 116.48,
        "deepswe": null,
        "scicode": 13.1944444444444,
        "aime": 1.6666666666666667,
        "gpqa": 26.957070707070706,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.0605,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "meta-llama/llama-3.2-1b-instruct",
      "label": "Meta: Llama 3.2 1B Instruct",
      "organization": "Meta AI",
      "released": "2024-09-24",
      "scores": {
        "eci": 101.95,
        "deepswe": null,
        "scicode": null,
        "aime": 0.5555555555555556,
        "gpqa": 23.92676767676768,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.08789999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "meta-llama/llama-3.3-70b-instruct",
      "label": "Meta: Llama 3.3 70B Instruct",
      "organization": "Meta AI",
      "released": "2024-12-06",
      "scores": {
        "eci": 127.32,
        "deepswe": null,
        "scicode": 26.041666666666703,
        "aime": 5.138888888888888,
        "gpqa": 47.44318181818182,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.177,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "meta-llama/llama-4-maverick",
      "label": "Meta: Llama 4 Maverick",
      "organization": "Meta AI",
      "released": "2025-04-06",
      "scores": {
        "eci": 132.2,
        "deepswe": null,
        "scicode": 33.1018518518519,
        "aime": 20.555555555555554,
        "gpqa": 66.98232323232324,
        "arc_agi_2": 0
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.35024999999999995,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "meta-llama/llama-4-scout",
      "label": "Meta: Llama 4 Scout",
      "organization": "Meta AI",
      "released": "2025-04-05",
      "scores": {
        "eci": 129.64,
        "deepswe": null,
        "scicode": 17.0138888888889,
        "aime": 7.777777777777778,
        "gpqa": 51.830808080808076,
        "arc_agi_2": 0
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.16999999999999998,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "meta/muse-spark-1.1",
      "label": "Meta: Muse Spark 1.1",
      "organization": "Meta AI",
      "released": "2026-07-09",
      "scores": {
        "eci": 154.32,
        "deepswe": 53.31858407079646,
        "scicode": 58.796296296296305,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": 2.361142644469026,
      "cost_per_1m_usd": 2.3,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "meta/muse-spark-1.2",
      "label": "Meta: Muse Spark 1.2",
      "organization": "Meta AI",
      "released": "2026-08-05",
      "scores": {
        "eci": 155.18,
        "deepswe": 54.86725663716814,
        "scicode": 56.3657407407407,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": 3.6955048180309733,
      "cost_per_1m_usd": 2.3,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "meta/muse-spark-1.3",
      "label": "Meta: Muse Spark 1.3",
      "organization": "Meta AI",
      "released": "2026-09-02",
      "scores": {
        "eci": 156.89,
        "deepswe": null,
        "scicode": 59.7222222222222,
        "aime": 99.16666666666667,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 2.3,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "microsoft/phi-4",
      "label": "Microsoft: Phi 4",
      "organization": "Microsoft Research",
      "released": "2024-12-12",
      "scores": {
        "eci": 130.43,
        "deepswe": null,
        "scicode": null,
        "aime": 13.750000000000002,
        "gpqa": 56.060606060606055,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.0945,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "MiMo-V2-Flash",
      "organization": "Xiaomi Corp",
      "released": "2025-12-16",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 25.925925925925903,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "minimax/minimax-m2.5",
      "label": "MiniMax: MiniMax M2.5",
      "organization": "MiniMax",
      "released": "2026-02-12",
      "scores": {
        "eci": 146.68,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": 4.859999999999999
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.5535,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "minimax/minimax-m2.7",
      "label": "MiniMax: MiniMax M2.7",
      "organization": "MiniMax",
      "released": "2026-03-18",
      "scores": {
        "eci": 145.84,
        "deepswe": null,
        "scicode": 46.9907407407407,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.4305,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "minimax/minimax-m3",
      "label": "MiniMax: MiniMax M3",
      "organization": "MiniMax",
      "released": "2026-06-01",
      "scores": {
        "eci": 147,
        "deepswe": null,
        "scicode": 47.1064814814815,
        "aime": 71.11111111111111,
        "gpqa": 90.9090909090909,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.615,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Ministral 3B",
      "organization": "Mistral AI",
      "released": "2024-10-16",
      "scores": {
        "eci": 118.08,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 25.252525252525253,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Ministral 8B",
      "organization": "Mistral AI",
      "released": "2024-10-16",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 27.146464646464647,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Mistral 7B v0.1",
      "organization": "Mistral AI",
      "released": "2023-09-27",
      "scores": {
        "eci": 111.97,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Mistral 7B v0.3",
      "organization": "Mistral AI",
      "released": "2024-05-27",
      "scores": {
        "eci": 108.73,
        "deepswe": null,
        "scicode": null,
        "aime": 0.2777777777777778,
        "gpqa": 15.18308080808081,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "mistralai/mistral-large",
      "label": "Mistral Large",
      "organization": "Mistral AI",
      "released": "2024-02-26",
      "scores": {
        "eci": 122,
        "deepswe": null,
        "scicode": null,
        "aime": 1.9444444444444444,
        "gpqa": 38.76262626262626,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 3.3999999999999995,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Mistral Large 2 (Nov 2024)",
      "organization": "Mistral AI",
      "released": "2024-07-24",
      "scores": {
        "eci": 127.54,
        "deepswe": null,
        "scicode": null,
        "aime": 8.472222222222223,
        "gpqa": 49.02146464646464,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Mistral Large 3",
      "organization": "Mistral AI",
      "released": "2025-12-02",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 36.2268518518519,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Mistral Small 3.1",
      "organization": "Mistral AI",
      "released": "2025-03-17",
      "scores": {
        "eci": 127.48,
        "deepswe": null,
        "scicode": 26.5046296296296,
        "aime": 5.833333333333333,
        "gpqa": 47.474747474747474,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Mistral Small 3.2",
      "organization": "Mistral AI",
      "released": "2025-06-20",
      "scores": {
        "eci": 131.73,
        "deepswe": null,
        "scicode": 26.3888888888889,
        "aime": 30.277777777777775,
        "gpqa": 49.053030303030305,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "mistralai/mistral-medium-3",
      "label": "Mistral: Mistral Medium 3",
      "organization": "Mistral AI",
      "released": "2025-05-07",
      "scores": {
        "eci": 134.07,
        "deepswe": null,
        "scicode": null,
        "aime": 32.22222222222222,
        "gpqa": 59.53282828282829,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.96,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "mistralai/mistral-medium-3.1",
      "label": "Mistral: Mistral Medium 3.1",
      "organization": "Mistral AI",
      "released": "2025-08-12",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 33.796296296296305,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.96,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "mistralai/mistral-medium-3-5",
      "label": "Mistral: Mistral Medium 3.5",
      "organization": "Mistral AI",
      "released": "2026-04-28",
      "scores": {
        "eci": 141.42,
        "deepswe": null,
        "scicode": 40.162037037037,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 3.6,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "mistralai/mistral-nemo",
      "label": "Mistral: Mistral Nemo",
      "organization": "Mistral AI",
      "released": "2024-07-18",
      "scores": {
        "eci": 118.63,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 29.892676767676768,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.022850000000000002,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "mistralai/mistral-small-24b-instruct-2501",
      "label": "Mistral: Mistral Small 3",
      "organization": "Mistral AI",
      "released": "2025-01-30",
      "scores": {
        "eci": 127.06,
        "deepswe": null,
        "scicode": null,
        "aime": 6.666666666666667,
        "gpqa": 47.285353535353536,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.0605,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "mistralai/mixtral-8x22b-instruct",
      "label": "Mistral: Mixtral 8x22B Instruct",
      "organization": "Mistral AI",
      "released": "2024-04-17",
      "scores": {
        "eci": 122,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 34.05934343434344,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 3.3999999999999995,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Mixtral 8x7B",
      "organization": "Mistral AI",
      "released": "2023-12-11",
      "scores": {
        "eci": 118.38,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 30.58712121212121,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "moonshotai/kimi-k2",
      "label": "MoonshotAI: Kimi K2 0711",
      "organization": "Moonshot",
      "released": "2025-07-12",
      "scores": {
        "eci": 140.13,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.1755,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "moonshotai/kimi-k2.5",
      "label": "MoonshotAI: Kimi K2.5",
      "organization": "Moonshot",
      "released": "2026-01-27",
      "scores": {
        "eci": 148.05,
        "deepswe": null,
        "scicode": 48.9583333333333,
        "aime": 92.2,
        "gpqa": 87.6,
        "arc_agi_2": 11.81
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.08,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "moonshotai/kimi-k2.6",
      "label": "MoonshotAI: Kimi K2.6",
      "organization": "Moonshot",
      "released": "2026-04-20",
      "scores": {
        "eci": 151.03,
        "deepswe": null,
        "scicode": 53.4722222222222,
        "aime": 96.11111111111111,
        "gpqa": 90.78282828282829,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.616,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "moonshotai/kimi-k2.7-code",
      "label": "MoonshotAI: Kimi K2.7 Code",
      "organization": "Moonshot",
      "released": "2026-06-12",
      "scores": {
        "eci": 150.01,
        "deepswe": 30.53097345132743,
        "scicode": 47.453703703703695,
        "aime": 95.55555555555556,
        "gpqa": 87.87878787878788,
        "arc_agi_2": null
      },
      "deepswe_task_usd": 2.815536243274336,
      "cost_per_1m_usd": 1.6087799999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "moonshotai/kimi-k3",
      "label": "MoonshotAI: Kimi K3",
      "organization": "Moonshot",
      "released": "2026-07-16",
      "scores": {
        "eci": 157.68,
        "deepswe": 68.51441241685144,
        "scicode": 59.4907407407407,
        "aime": 97.22222222222221,
        "gpqa": 93.11868686868688,
        "arc_agi_2": 60.416666666666664
      },
      "deepswe_task_usd": 4.654682129933482,
      "cost_per_1m_usd": 7.2,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "MPT-30B",
      "organization": "MosaicML",
      "released": "2023-06-22",
      "scores": {
        "eci": 100.16,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "MPT-7B",
      "organization": "MosaicML",
      "released": "2023-05-05",
      "scores": {
        "eci": 93.97,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Muse Spark",
      "organization": "Meta AI",
      "released": "2026-04-08",
      "scores": {
        "eci": 152.09,
        "deepswe": null,
        "scicode": 51.5046296296296,
        "aime": 88.9,
        "gpqa": 89.8,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Nemotron-4 15B",
      "organization": "NVIDIA",
      "released": "2024-02-26",
      "scores": {
        "eci": 107.38,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Nova 2.0 Pro Preview",
      "organization": "Amazon",
      "released": "2025-12-02",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 42.7083333333333,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "nvidia/nemotron-3-super-120b-a12b",
      "label": "NVIDIA: Nemotron 3 Super",
      "organization": "Nvidia",
      "released": "2026-03-11",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 35.9953703703704,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.20949999999999996,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "nvidia/nemotron-3-ultra-550b-a55b",
      "label": "NVIDIA: Nemotron 3 Ultra",
      "organization": "NVIDIA",
      "released": "2026-06-04",
      "scores": {
        "eci": 146.2,
        "deepswe": null,
        "scicode": 40.2777777777778,
        "aime": 86.66666666666667,
        "gpqa": 85.35353535353535,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.23,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "o1-mini",
      "organization": "OpenAI",
      "released": "2024-09-12",
      "scores": {
        "eci": 135.82,
        "deepswe": null,
        "scicode": null,
        "aime": 46.94444444444444,
        "gpqa": 62.37373737373737,
        "arc_agi_2": 0.83
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "open_llama_7b",
      "organization": null,
      "released": "2023-06-07",
      "scores": {
        "eci": 90.96,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-3.5-turbo",
      "label": "OpenAI: GPT-3.5 Turbo",
      "organization": "OpenAI",
      "released": "2023-06-13",
      "scores": {
        "eci": 113.14,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.8499999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-4",
      "label": "OpenAI: GPT-4",
      "organization": "OpenAI",
      "released": "2023-06-13",
      "scores": {
        "eci": 123.1,
        "deepswe": null,
        "scicode": null,
        "aime": 1.1111111111111112,
        "gpqa": 30.650252525252526,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 40.5,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-4-turbo",
      "label": "OpenAI: GPT-4 Turbo",
      "organization": "OpenAI",
      "released": "2024-01-25",
      "scores": {
        "eci": 126.46,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 42.36111111111111,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 17,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-4.1",
      "label": "OpenAI: GPT-4.1",
      "organization": "OpenAI",
      "released": "2025-04-14",
      "scores": {
        "eci": 136.79,
        "deepswe": null,
        "scicode": null,
        "aime": 38.333333333333336,
        "gpqa": 66.91919191919192,
        "arc_agi_2": 0.42
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 4.1,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-4.1-mini",
      "label": "OpenAI: GPT-4.1 Mini",
      "organization": "OpenAI",
      "released": "2025-04-14",
      "scores": {
        "eci": 135.01,
        "deepswe": null,
        "scicode": 40.3935185185185,
        "aime": 44.72222222222222,
        "gpqa": 65.84595959595958,
        "arc_agi_2": 0
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.82,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-4.1-nano",
      "label": "OpenAI: GPT-4.1 Nano",
      "organization": "OpenAI",
      "released": "2025-04-14",
      "scores": {
        "eci": 129.62,
        "deepswe": null,
        "scicode": 25.925925925925903,
        "aime": 28.888888888888886,
        "gpqa": 48.926767676767675,
        "arc_agi_2": 0
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.205,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-4o",
      "label": "OpenAI: GPT-4o",
      "organization": "OpenAI",
      "released": "2024-08-06",
      "scores": {
        "eci": 128.77,
        "deepswe": null,
        "scicode": null,
        "aime": 6.388888888888888,
        "gpqa": 49.21085858585859,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 5.125,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-4o-mini",
      "label": "OpenAI: GPT-4o-mini",
      "organization": "OpenAI",
      "released": "2024-07-18",
      "scores": {
        "eci": 126.56,
        "deepswe": null,
        "scicode": null,
        "aime": 6.944444444444445,
        "gpqa": 37.7209595959596,
        "arc_agi_2": 0
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.3075,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5",
      "label": "OpenAI: GPT-5",
      "organization": "OpenAI",
      "released": "2025-08-07",
      "scores": {
        "eci": 150,
        "deepswe": null,
        "scicode": 42.939814814814795,
        "aime": 91.38888888888889,
        "gpqa": 86.17424242424242,
        "arc_agi_2": 9.86
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 4.3125,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5-mini",
      "label": "OpenAI: GPT-5 Mini",
      "organization": "OpenAI",
      "released": "2025-08-07",
      "scores": {
        "eci": 145.51,
        "deepswe": null,
        "scicode": 39.2361111111111,
        "aime": 86.66666666666667,
        "gpqa": 75,
        "arc_agi_2": 4.44
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.8624999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5-nano",
      "label": "OpenAI: GPT-5 Nano",
      "organization": "OpenAI",
      "released": "2025-08-07",
      "scores": {
        "eci": 139.39,
        "deepswe": null,
        "scicode": null,
        "aime": 81.11111111111111,
        "gpqa": 69.44444444444444,
        "arc_agi_2": 2.6100000000000003
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.1725,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5-pro",
      "label": "OpenAI: GPT-5 Pro",
      "organization": "OpenAI",
      "released": "2025-10-07",
      "scores": {
        "eci": 150.26,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": 18.33
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 51.75,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.1",
      "label": "OpenAI: GPT-5.1",
      "organization": "OpenAI",
      "released": "2025-11-13",
      "scores": {
        "eci": 149.66,
        "deepswe": null,
        "scicode": 43.287037037037,
        "aime": 88.61111111111111,
        "gpqa": 87.62626262626263,
        "arc_agi_2": 17.64
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 4.3125,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.2",
      "label": "OpenAI: GPT-5.2",
      "organization": "OpenAI",
      "released": "2025-12-11",
      "scores": {
        "eci": 153.45,
        "deepswe": null,
        "scicode": null,
        "aime": 96.11111111111111,
        "gpqa": 91.4,
        "arc_agi_2": 52.910000000000004
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 6.0375,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.2-pro",
      "label": "OpenAI: GPT-5.2 Pro",
      "organization": "OpenAI",
      "released": "2025-12-11",
      "scores": {
        "eci": 155.41,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": 54.16
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 72.45,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.3-codex",
      "label": "OpenAI: GPT-5.3-Codex",
      "organization": "OpenAI",
      "released": "2026-02-05",
      "scores": {
        "eci": 156.84,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 6.0375,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.4",
      "label": "OpenAI: GPT-5.4",
      "organization": "OpenAI",
      "released": "2026-03-05",
      "scores": {
        "eci": 156.92,
        "deepswe": 51.76991150442478,
        "scicode": 56.5972222222222,
        "aime": 97.77777777777777,
        "gpqa": 93.30000000000001,
        "arc_agi_2": 73.95
      },
      "deepswe_task_usd": 5.6524577245575225,
      "cost_per_1m_usd": 6.875,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.4-mini",
      "label": "OpenAI: GPT-5.4 Mini",
      "organization": "OpenAI",
      "released": "2026-03-17",
      "scores": {
        "eci": 148.83,
        "deepswe": null,
        "scicode": 49.88425925925931,
        "aime": 88.88888888888889,
        "gpqa": 86.86868686868688,
        "arc_agi_2": 18.9
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 2.0625,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.4-nano",
      "label": "OpenAI: GPT-5.4 Nano",
      "organization": "OpenAI",
      "released": "2026-03-17",
      "scores": {
        "eci": 145.78,
        "deepswe": null,
        "scicode": 46.875,
        "aime": 87.78,
        "gpqa": 78.47,
        "arc_agi_2": 5.6899999999999995
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.5675,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.4-pro",
      "label": "OpenAI: GPT-5.4 Pro",
      "organization": "OpenAI",
      "released": "2026-03-05",
      "scores": {
        "eci": 159.12,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 94.6,
        "arc_agi_2": 83.33
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 82.5,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.5",
      "label": "OpenAI: GPT-5.5",
      "organization": "OpenAI",
      "released": "2026-04-23",
      "scores": {
        "eci": 159.25,
        "deepswe": 67.03539823008849,
        "scicode": 56.1342592592593,
        "aime": 100,
        "gpqa": 94.00252525252525,
        "arc_agi_2": 85
      },
      "deepswe_task_usd": 7.226236674778761,
      "cost_per_1m_usd": 13.75,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.5-pro",
      "label": "OpenAI: GPT-5.5 Pro",
      "organization": "OpenAI",
      "released": "2026-04-23",
      "scores": {
        "eci": 162.45,
        "deepswe": null,
        "scicode": null,
        "aime": 100,
        "gpqa": 93.92135642135642,
        "arc_agi_2": 84.58
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 82.5,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.6-luna",
      "label": "OpenAI: GPT-5.6 Luna",
      "organization": "OpenAI",
      "released": "2026-07-09",
      "scores": {
        "eci": 156.32,
        "deepswe": 67.1875,
        "scicode": 53.587962962963,
        "aime": 98.33333333333333,
        "gpqa": 91.60353535353535,
        "arc_agi_2": 59.540000000000006
      },
      "deepswe_task_usd": 3.028116810267857,
      "cost_per_1m_usd": 0.55,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.6-sol",
      "label": "OpenAI: GPT-5.6 Sol",
      "organization": "OpenAI",
      "released": "2026-07-09",
      "scores": {
        "eci": 161.99,
        "deepswe": 72.66666666666667,
        "scicode": 57.0601851851852,
        "aime": 100,
        "gpqa": 93.49747474747475,
        "arc_agi_2": 92.5
      },
      "deepswe_task_usd": 8.386436346666667,
      "cost_per_1m_usd": 4.8,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-5.6-terra",
      "label": "OpenAI: GPT-5.6 Terra",
      "organization": "OpenAI",
      "released": "2026-07-09",
      "scores": {
        "eci": 159.31,
        "deepswe": 69.62305986696231,
        "scicode": 54.976851851851805,
        "aime": 99.72222222222223,
        "gpqa": 93.3080808080808,
        "arc_agi_2": 83.89999999999999
      },
      "deepswe_task_usd": 4.945847263858093,
      "cost_per_1m_usd": 5.499999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-6-astra",
      "label": "OpenAI: GPT-6 Astra",
      "organization": "OpenAI",
      "released": "2026-09-03",
      "scores": {
        "eci": 166.6,
        "deepswe": 74.11504424778761,
        "scicode": 56.4814814814815,
        "aime": 100,
        "gpqa": 95.77020202020202,
        "arc_agi_2": 95
      },
      "deepswe_task_usd": 6.52377356460177,
      "cost_per_1m_usd": 24,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-6-luna",
      "label": "OpenAI: GPT-6 Luna",
      "organization": "OpenAI",
      "released": "2026-09-22",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.24,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-6-sol",
      "label": "OpenAI: GPT-6 Sol",
      "organization": "OpenAI",
      "released": "2026-09-22",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 4.8,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-oss-120b",
      "label": "OpenAI: gpt-oss-120b",
      "organization": "OpenAI",
      "released": "2025-08-05",
      "scores": {
        "eci": 139.94,
        "deepswe": null,
        "scicode": 35.9953703703704,
        "aime": 88.88888888888889,
        "gpqa": 75.75757575757575,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.08354999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/gpt-oss-20b",
      "label": "OpenAI: gpt-oss-20b",
      "organization": "OpenAI",
      "released": "2025-08-05",
      "scores": {
        "eci": 137.81,
        "deepswe": null,
        "scicode": 34.375,
        "aime": 65.27777777777779,
        "gpqa": 60.79545454545454,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.0432,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/o1",
      "label": "OpenAI: o1",
      "organization": "OpenAI",
      "released": "2024-09-12",
      "scores": {
        "eci": 134.78,
        "deepswe": null,
        "scicode": null,
        "aime": 31.11111111111111,
        "gpqa": 50.31565656565656,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 30.75,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/o3",
      "label": "OpenAI: o3",
      "organization": "OpenAI",
      "released": "2025-04-16",
      "scores": {
        "eci": 146.88,
        "deepswe": null,
        "scicode": null,
        "aime": 84.44444444444444,
        "gpqa": 81.81818181818183,
        "arc_agi_2": 6.529999999999999
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 4.1,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/o3-mini",
      "label": "OpenAI: o3 Mini",
      "organization": "OpenAI",
      "released": "2025-01-31",
      "scores": {
        "eci": 140.35,
        "deepswe": null,
        "scicode": 39.785641909282496,
        "aime": 76.94444444444444,
        "gpqa": 77.02020202020202,
        "arc_agi_2": 2.9899999999999998
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 2.255,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/o3-pro",
      "label": "OpenAI: o3 Pro",
      "organization": "OpenAI",
      "released": "2025-06-10",
      "scores": {
        "eci": 147.43,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": 4.859999999999999
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 41,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "openai/o4-mini",
      "label": "OpenAI: o4 Mini",
      "organization": "OpenAI",
      "released": "2025-04-16",
      "scores": {
        "eci": 145.63,
        "deepswe": null,
        "scicode": null,
        "aime": 81.66666666666667,
        "gpqa": 79.60858585858585,
        "arc_agi_2": 6.11
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 2.255,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "PaLM 2-L",
      "organization": null,
      "released": "2023-05-17",
      "scores": {
        "eci": 114.84,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "PaLM 2-M",
      "organization": null,
      "released": "2023-05-17",
      "scores": {
        "eci": 107.96,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "PaLM 2-S",
      "organization": null,
      "released": "2023-05-17",
      "scores": {
        "eci": 105.79,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Phi-1.5",
      "organization": "Microsoft",
      "released": "2023-09-11",
      "scores": {
        "eci": 90.82,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Phi-2",
      "organization": "Microsoft",
      "released": "2023-12-12",
      "scores": {
        "eci": 107.62,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "phi-3-medium 14B",
      "organization": "Microsoft",
      "released": "2024-04-23",
      "scores": {
        "eci": 121.18,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 27.58838383838384,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "phi-3-mini 3.8B",
      "organization": "Microsoft",
      "released": "2024-04-23",
      "scores": {
        "eci": 117.23,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "phi-3-small 7.4B",
      "organization": "Microsoft",
      "released": "2024-04-23",
      "scores": {
        "eci": 121.79,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Phi-4 Mini",
      "organization": "Microsoft",
      "released": "2025-02-27",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 10.7638888888889,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen 3.5 Flash (hosted 35B-A3B)",
      "organization": "Alibaba",
      "released": "2026-02-25",
      "scores": {
        "eci": 143.98,
        "deepswe": null,
        "scicode": null,
        "aime": 84.44444444444444,
        "gpqa": 82.32323232323232,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen 3.5 Plus (hosted 397B-A17B)",
      "organization": "Alibaba",
      "released": "2026-02-16",
      "scores": {
        "eci": 146.79,
        "deepswe": null,
        "scicode": null,
        "aime": 86.66666666666667,
        "gpqa": 84.84848484848484,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen 3.6 35B-A3B",
      "organization": "Alibaba",
      "released": "2026-04-14",
      "scores": {
        "eci": 143.92,
        "deepswe": null,
        "scicode": 35.7638888888889,
        "aime": 86.66666666666667,
        "gpqa": 84.84848484848484,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen 3.6 Flash",
      "organization": "Alibaba",
      "released": "2026-04-27",
      "scores": {
        "eci": 143.26,
        "deepswe": null,
        "scicode": null,
        "aime": 84.44444444444444,
        "gpqa": 83.33333333333334,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen 3.6 Max (Preview)",
      "organization": "Alibaba",
      "released": "2026-04-20",
      "scores": {
        "eci": 149.27,
        "deepswe": null,
        "scicode": null,
        "aime": 91.11111111111111,
        "gpqa": 87.37373737373737,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen 3.6 Plus",
      "organization": "Alibaba",
      "released": "2026-03-31",
      "scores": {
        "eci": 147.67,
        "deepswe": null,
        "scicode": 40.7407407407407,
        "aime": 93.33333333333333,
        "gpqa": 88.38383838383838,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen 3.8 27B",
      "organization": "Alibaba",
      "released": "2026-08-14",
      "scores": {
        "eci": 149.38,
        "deepswe": null,
        "scicode": 46.6435185185185,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen 3.8 Max",
      "organization": "Alibaba",
      "released": "2026-08-02",
      "scores": {
        "eci": 156.69,
        "deepswe": 57.46102449888642,
        "scicode": 52.8935185185185,
        "aime": 99.44444444444444,
        "gpqa": 92.67676767676768,
        "arc_agi_2": null
      },
      "deepswe_task_usd": 3.7290646500445437,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen-1_8B",
      "organization": null,
      "released": "2023-11-30",
      "scores": {
        "eci": 92.28,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen-14B",
      "organization": "Alibaba",
      "released": "2023-09-24",
      "scores": {
        "eci": 112.8,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen-7B",
      "organization": "Alibaba",
      "released": "2023-09-28",
      "scores": {
        "eci": 106.47,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen-Turbo",
      "organization": "Alibaba",
      "released": "2024-11-01",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 6.111111111111111,
        "gpqa": 41.792929292929294,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen-plus",
      "label": "Qwen: Qwen-Plus",
      "organization": "Alibaba",
      "released": "2025-01-25",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 17.77777777777778,
        "gpqa": 48.10606060606061,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.44199999999999995,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen-2.5-7b-instruct",
      "label": "Qwen: Qwen2.5 7B Instruct",
      "organization": "Alibaba",
      "released": "2024-09-19",
      "scores": {
        "eci": 118.44,
        "deepswe": null,
        "scicode": null,
        "aime": 2.5,
        "gpqa": 35.47979797979798,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.135,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3-14b",
      "label": "Qwen: Qwen3 14B",
      "organization": "Alibaba",
      "released": "2025-04-29",
      "scores": {
        "eci": 138.23,
        "deepswe": null,
        "scicode": 31.5972222222222,
        "aime": 66.38888888888889,
        "gpqa": 63.76262626262626,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.16199999999999998,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3-235b-a22b",
      "label": "Qwen: Qwen3 235B A22B",
      "organization": "Alibaba",
      "released": "2025-07-25",
      "scores": {
        "eci": 138.93,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": 1.25
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.9327499999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3-30b-a3b",
      "label": "Qwen: Qwen3 30B A3B",
      "organization": "Alibaba",
      "released": "2025-04-29",
      "scores": {
        "eci": 136.18,
        "deepswe": null,
        "scicode": null,
        "aime": 62.77777777777778,
        "gpqa": 61.67929292929293,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.2665,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3-32b",
      "label": "Qwen: Qwen3 32B",
      "organization": "Alibaba",
      "released": "2025-04-29",
      "scores": {
        "eci": 138.51,
        "deepswe": null,
        "scicode": 35.4166666666667,
        "aime": 66.94444444444444,
        "gpqa": 65.71969696969697,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.15000000000000002,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3-8b",
      "label": "Qwen: Qwen3 8B",
      "organization": "Alibaba",
      "released": "2025-04-28",
      "scores": {
        "eci": 136.17,
        "deepswe": null,
        "scicode": 22.5694444444444,
        "aime": 56.111111111111114,
        "gpqa": 56.755050505050505,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.23529999999999998,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3-coder-next",
      "label": "Qwen: Qwen3 Coder Next",
      "organization": null,
      "released": "2026-02-02",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 32.2916666666667,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.358,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3-max",
      "label": "Qwen: Qwen3 Max",
      "organization": "Alibaba",
      "released": "2025-09-24",
      "scores": {
        "eci": 142.41,
        "deepswe": null,
        "scicode": null,
        "aime": 73.33333333333333,
        "gpqa": 72.6010101010101,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.8719999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3.5-397b-a17b",
      "label": "Qwen: Qwen3.5 397B A17B",
      "organization": "Alibaba",
      "released": "2026-02-13",
      "scores": {
        "eci": 146.64,
        "deepswe": null,
        "scicode": null,
        "aime": 88.88888888888889,
        "gpqa": 86.36363636363636,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.5825,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3.5-122b-a10b",
      "label": "Qwen: Qwen3.5-122B-A10B",
      "organization": "Alibaba",
      "released": "2026-02-24",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 35.6481481481482,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.897,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3.5-35b-a3b",
      "label": "Qwen: Qwen3.5-35B-A3B",
      "organization": "Alibaba",
      "released": "2026-02-24",
      "scores": {
        "eci": 142.52,
        "deepswe": null,
        "scicode": 29.2824074074074,
        "aime": 70,
        "gpqa": 83.45959595959596,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.5606249999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3.5-9b",
      "label": "Qwen: Qwen3.5-9B",
      "organization": "Alibaba",
      "released": "2026-02-24",
      "scores": {
        "eci": 139.45,
        "deepswe": null,
        "scicode": 27.5462962962963,
        "aime": 61.66666666666667,
        "gpqa": 78.97727272727273,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.1175,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3.6-27b",
      "label": "Qwen: Qwen3.6 27B",
      "organization": "Alibaba",
      "released": "2026-04-22",
      "scores": {
        "eci": 146.47,
        "deepswe": null,
        "scicode": 37.2685185185185,
        "aime": 91.11111111111111,
        "gpqa": 85.85858585858585,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.3279999999999998,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3.7-flash",
      "label": "Qwen: Qwen3.7 Flash",
      "organization": "Alibaba",
      "released": "2026-07-27",
      "scores": {
        "eci": 144.62,
        "deepswe": null,
        "scicode": null,
        "aime": 86.66666666666667,
        "gpqa": 82.32323232323232,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.065,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3.7-max",
      "label": "Qwen: Qwen3.7 Max",
      "organization": "Alibaba",
      "released": "2026-05-19",
      "scores": {
        "eci": 153.74,
        "deepswe": null,
        "scicode": 48.842592592592595,
        "aime": 95.55555555555556,
        "gpqa": 90.9090909090909,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 2.5075,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3.7-plus",
      "label": "Qwen: Qwen3.7 Plus",
      "organization": "Alibaba",
      "released": "2026-06-02",
      "scores": {
        "eci": 147.39,
        "deepswe": null,
        "scicode": 45.4861111111111,
        "aime": 93.33333333333333,
        "gpqa": 87.87878787878788,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.6559999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3.8-flash",
      "label": "Qwen: Qwen3.8 Flash",
      "organization": "Alibaba",
      "released": "2026-08-26",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.262,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen3.8-max-0902",
      "label": "Qwen: Qwen3.8 Max (0902)",
      "organization": "Alibaba",
      "released": "2026-09-01",
      "scores": {
        "eci": 155.28,
        "deepswe": null,
        "scicode": 52.0833333333333,
        "aime": 100,
        "gpqa": 92.2979797979798,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 3.3999999999999995,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen1.5-32B",
      "organization": "Alibaba",
      "released": "2024-04-03",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 30.7449494949495,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen1.5-72B",
      "organization": "Alibaba",
      "released": "2024-02-04",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 28.819444444444443,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen2-72B",
      "organization": "Alibaba",
      "released": "2024-06-07",
      "scores": {
        "eci": 125.27,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 40.782828282828284,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen-2.5-72b-instruct",
      "label": "Qwen2.5 72B Instruct",
      "organization": "Alibaba",
      "released": "2024-09-19",
      "scores": {
        "eci": 129.01,
        "deepswe": null,
        "scicode": null,
        "aime": 8.055555555555555,
        "gpqa": 49.14772727272727,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.374,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "qwen/qwen-2.5-coder-32b-instruct",
      "label": "Qwen2.5 Coder 32B Instruct",
      "organization": "Alibaba",
      "released": "2024-09-18",
      "scores": {
        "eci": 119.41,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.779,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen2.5-32B",
      "organization": "Alibaba",
      "released": "2024-09-17",
      "scores": {
        "eci": 128.52,
        "deepswe": null,
        "scicode": null,
        "aime": 7.361111111111112,
        "gpqa": 46.08585858585859,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen2.5-Coder (7B)",
      "organization": "Alibaba",
      "released": "2024-09-18",
      "scores": {
        "eci": 102.56,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen2.5-Coder-0.5B",
      "organization": null,
      "released": "2024-09-18",
      "scores": {
        "eci": 87.38,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen2.5-Coder-14B",
      "organization": null,
      "released": "2024-09-18",
      "scores": {
        "eci": 116.26,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen2.5-Coder-3B",
      "organization": null,
      "released": "2024-09-18",
      "scores": {
        "eci": 107.4,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen2.5-Max",
      "organization": "Alibaba",
      "released": "2025-01-25",
      "scores": {
        "eci": 132.54,
        "deepswe": null,
        "scicode": null,
        "aime": 16.11111111111111,
        "gpqa": 56.12373737373737,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen3-1.7B",
      "organization": "Alibaba",
      "released": "2025-04-29",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 8.055555555555555,
        "gpqa": 38.005050505050505,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen3-4B",
      "organization": "Alibaba",
      "released": "2025-04-29",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 52.33585858585859,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "qwen3-4b-instruct-2507",
      "organization": null,
      "released": "2025-08-06",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 52.22222222222223,
        "gpqa": 45.83333333333333,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Qwen3.5-4B",
      "organization": "Alibaba",
      "released": "2026-02-24",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 55.833333333333336,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "QwQ-32B",
      "organization": "Alibaba",
      "released": "2025-03-05",
      "scores": {
        "eci": 137.6,
        "deepswe": null,
        "scicode": null,
        "aime": 59.166666666666664,
        "gpqa": 65.3409090909091,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "QWQ-Plus",
      "organization": "Alibaba",
      "released": "2025-04-08",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 65.40404040404042,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "RedPajama-INCITE-7B-Base",
      "organization": null,
      "released": "2023-05-04",
      "scores": {
        "eci": 89.71,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Ring-2.6-1T",
      "organization": "Ant Group",
      "released": "2026-05-14",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 42.3611111111111,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "seed-oss-36b-instruct",
      "organization": null,
      "released": "2025-08-20",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 67.5,
        "gpqa": 71.52777777777779,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "x-ai/grok-4.20",
      "label": "SpaceXAI: Grok 4.20",
      "organization": "xAI",
      "released": "2026-02-17",
      "scores": {
        "eci": 152.01,
        "deepswe": null,
        "scicode": null,
        "aime": 92.22222222222223,
        "gpqa": 89.33080808080808,
        "arc_agi_2": 65.14
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.6875,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "x-ai/grok-4.5",
      "label": "SpaceXAI: Grok 4.5",
      "organization": "xAI",
      "released": "2026-07-08",
      "scores": {
        "eci": 154.02,
        "deepswe": 53.76106194690266,
        "scicode": 54.050925925925895,
        "aime": 97.77777777777777,
        "gpqa": 93.43434343434343,
        "arc_agi_2": 52.64
      },
      "deepswe_task_usd": 2.4157481725663716,
      "cost_per_1m_usd": 3.3999999999999995,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "x-ai/grok-4.6",
      "label": "SpaceXAI: Grok 4.6",
      "organization": "xAI",
      "released": "2026-08-12",
      "scores": {
        "eci": 156.48,
        "deepswe": 67.47787610619469,
        "scicode": 56.4814814814815,
        "aime": 99.16666666666667,
        "gpqa": 94.00252525252525,
        "arc_agi_2": 67.08333333333333
      },
      "deepswe_task_usd": 3.4489837522123894,
      "cost_per_1m_usd": 3.3999999999999995,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "x-ai/grok-4.7",
      "label": "SpaceXAI: Grok 4.7",
      "organization": "xAI",
      "released": "2026-09-21",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 57.4074074074074,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 3.3999999999999995,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "x-ai/grok-build-0.1",
      "label": "SpaceXAI: Grok Build 0.1",
      "organization": "xAI",
      "released": "2026-05-29",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 50.2314814814815,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.35,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Stable Beluga 2",
      "organization": "Stability AI",
      "released": "2023-07-20",
      "scores": {
        "eci": 116.97,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "stablelm-tuned-alpha-7b",
      "organization": null,
      "released": "2023-04-19",
      "scores": {
        "eci": 54.28,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "StarCoder 2 15B",
      "organization": "Hugging Face,ServiceNow,NVIDIA,BigCode",
      "released": "2024-02-20",
      "scores": {
        "eci": 104.61,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "StarCoder 2 3B",
      "organization": "Hugging Face,ServiceNow,NVIDIA,BigCode",
      "released": "2024-02-22",
      "scores": {
        "eci": 88.01,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "StarCoder 2 7B",
      "organization": "Hugging Face,ServiceNow,NVIDIA,BigCode",
      "released": "2024-02-20",
      "scores": {
        "eci": 92.92,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Step 5 Preview",
      "organization": "StepFun",
      "released": "2026-09-18",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 58.912037037037,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "stepfun/step-3.7-flash",
      "label": "StepFun: Step 3.7 Flash",
      "organization": "StepFun",
      "released": "2026-05-29",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 40.046296296296305,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.5325,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "thinkingmachines/inkling",
      "label": "Thinking Machines: Inkling",
      "organization": "Thinking Machines",
      "released": "2026-07-15",
      "scores": {
        "eci": 148.6,
        "deepswe": null,
        "scicode": 46.9907407407407,
        "aime": 88.88888888888889,
        "gpqa": 88.25757575757575,
        "arc_agi_2": 36.52777777777778
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 2.0675,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "thinkingmachines/inkling-small",
      "label": "Thinking Machines: Inkling Small",
      "organization": "Thinking Machines",
      "released": "2026-07-15",
      "scores": {
        "eci": 150.13,
        "deepswe": null,
        "scicode": 48.7268518518519,
        "aime": 90,
        "gpqa": 88.51010101010101,
        "arc_agi_2": 40.138888888888886
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.7124999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "tiny-recursion-model",
      "organization": null,
      "released": "2025-10-06",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": 6.25
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Tulu 3 (Tülu 3) 70B",
      "organization": "Allen Institute for AI,University of Washington",
      "released": "2024-11-21",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 4.444444444444445,
        "gpqa": 46.275252525252526,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "upstage/solar-pro-3",
      "label": "Upstage: Solar Pro 3",
      "organization": "Upstage",
      "released": "2026-01-27",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 24.6527777777778,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.3075,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "vicuna-13b-v1.1",
      "organization": null,
      "released": "2023-04-12",
      "scores": {
        "eci": 94.04,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "microsoft/wizardlm-2-8x22b",
      "label": "WizardLM-2 8x22B",
      "organization": "Microsoft",
      "released": "2024-04-15",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 43.43434343434344,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.62,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "XGen-7B",
      "organization": "Salesforce",
      "released": "2023-06-27",
      "scores": {
        "eci": 92.75,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "xiaomi/mimo-v2.5",
      "label": "Xiaomi: MiMo-V2.5",
      "organization": "Xiaomi Corp",
      "released": "2026-04-23",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 43.0555555555556,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.189,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "xiaomi/mimo-v2.5-pro",
      "label": "Xiaomi: MiMo-V2.5-Pro",
      "organization": "Xiaomi Corp",
      "released": "2026-04-23",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 50.2314814814815,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.58725,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Yi 6B",
      "organization": "01.AI",
      "released": "2023-11-22",
      "scores": {
        "eci": 104.37,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Yi-1.5-34B",
      "organization": "01.AI",
      "released": "2024-05-13",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 31.9760101010101,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Yi-34B",
      "organization": "01.AI",
      "released": "2023-11-02",
      "scores": {
        "eci": 117.28,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": 14.741161616161616,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": null,
      "label": "Yi-9B",
      "organization": null,
      "released": "2024-03-01",
      "scores": {
        "eci": 107.27,
        "deepswe": null,
        "scicode": null,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": null,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "z-ai/glm-4.6",
      "label": "Z.ai: GLM 4.6",
      "organization": "Z.ai (Zhipu AI),Tsinghua University",
      "released": "2025-09-30",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": 38.4259259259259,
        "aime": null,
        "gpqa": null,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.8919999999999999,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "z-ai/glm-4.7",
      "label": "Z.ai: GLM 4.7",
      "organization": "Z.ai (Zhipu AI)",
      "released": "2025-12-22",
      "scores": {
        "eci": 143.53,
        "deepswe": null,
        "scicode": 45.1388888888889,
        "aime": 83.33333333333334,
        "gpqa": 83.33333333333334,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.1600000000000001,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "z-ai/glm-4.7-flash",
      "label": "Z.ai: GLM 4.7 Flash",
      "organization": null,
      "released": "2026-01-19",
      "scores": {
        "eci": null,
        "deepswe": null,
        "scicode": null,
        "aime": 58.333333333333336,
        "gpqa": 60.542929292929294,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 0.17932499999999998,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "z-ai/glm-5",
      "label": "Z.ai: GLM 5",
      "organization": "Z.ai (Zhipu AI)",
      "released": "2026-02-11",
      "scores": {
        "eci": 145.84,
        "deepswe": null,
        "scicode": null,
        "aime": 80,
        "gpqa": 87.81725888324873,
        "arc_agi_2": 4.859999999999999
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 1.0619999999999998,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "z-ai/glm-5.1",
      "label": "Z.ai: GLM 5.1",
      "organization": "Z.ai (Zhipu AI)",
      "released": "2026-04-07",
      "scores": {
        "eci": 149.86,
        "deepswe": null,
        "scicode": 43.75,
        "aime": 93.33333333333333,
        "gpqa": 89.8989898989899,
        "arc_agi_2": null
      },
      "deepswe_task_usd": null,
      "cost_per_1m_usd": 2.45,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "z-ai/glm-5.2",
      "label": "Z.ai: GLM 5.2",
      "organization": "Z.ai (Zhipu AI)",
      "released": "2026-06-16",
      "scores": {
        "eci": 151.76,
        "deepswe": 43.777777777777786,
        "scicode": 50.462962962963,
        "aime": 86.38888888888889,
        "gpqa": 91.85606060606061,
        "arc_agi_2": 22.775000000000002
      },
      "deepswe_task_usd": 3.9199075743111114,
      "cost_per_1m_usd": 1.559,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "z-ai/glm-5.3",
      "label": "Z.ai: GLM 5.3",
      "organization": "Z.ai (Zhipu AI)",
      "released": "2026-08-14",
      "scores": {
        "eci": 155.56,
        "deepswe": 68.95787139689578,
        "scicode": 59.0277777777778,
        "aime": 91.11111111111111,
        "gpqa": 90.9090909090909,
        "arc_agi_2": null
      },
      "deepswe_task_usd": 3.9933584893126386,
      "cost_per_1m_usd": 2.45,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    },
    {
      "model_id": "z-ai/glm-5.3-flash",
      "label": "Z.ai: GLM 5.3 Flash",
      "organization": "Z.ai (Zhipu AI)",
      "released": "2026-08-20",
      "scores": {
        "eci": 151.9,
        "deepswe": 63.39285714285714,
        "scicode": 51.6203703703704,
        "aime": 93.88888888888889,
        "gpqa": 90.15151515151516,
        "arc_agi_2": null
      },
      "deepswe_task_usd": 0.48196371245535713,
      "cost_per_1m_usd": 0.27249999999999996,
      "aa": null,
      "sources": [
        "https://epoch.ai/benchmarks"
      ]
    }
  ]
}
