{
  "generated": "2026-09-08",
  "benchmark": "mmlu",
  "results": [
    {
      "system": "GPT-3 175B (few-shot)",
      "developer": "OpenAI",
      "value": 43.9,
      "date": "2020-09-07",
      "source": {
        "url": "https://arxiv.org/abs/2009.03300",
        "kind": "paper",
        "accessed": "2026-09-04"
      },
      "conditions": {
        "shots": 5,
        "notes": "Table 1 of the MMLU paper (GPT-3 X-Large)."
      }
    },
    {
      "system": "gpt-4-turbo-2024-04-09",
      "developer": "OpenAI",
      "value": 86.7,
      "date": "2024-04-09",
      "source": {
        "url": "https://github.com/openai/simple-evals",
        "kind": "developer-report",
        "accessed": "2026-09-04"
      },
      "conditions": {
        "shots": 0,
        "notes": "simple-evals README (zero-shot chain-of-thought)."
      }
    },
    {
      "system": "o1",
      "developer": "OpenAI",
      "value": 91.8,
      "date": "2024-12-17",
      "source": {
        "url": "https://github.com/openai/simple-evals",
        "kind": "developer-report",
        "accessed": "2026-09-04"
      },
      "conditions": {
        "shots": 0,
        "notes": "simple-evals README; o3-high later scored 93.3 (April 2025)."
      }
    },
    {
      "system": "DeepSeek-V4-Pro-Base",
      "developer": "DeepSeek",
      "value": 90.1,
      "date": "2026-04-22",
      "source": {
        "url": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro",
        "kind": "developer-report",
        "accessed": "2026-09-04"
      },
      "conditions": {
        "shots": 5,
        "notes": "DeepSeek-V4 model card, base-model table (HF repo created 2026-04-22). Base (pre-trained) model, exact match."
      }
    }
  ]
}