{
  "generated": "2026-09-08",
  "benchmark": "tool-decathlon",
  "results": [
    {
      "system": "Claude-4.5-Sonnet",
      "developer": "Anthropic",
      "value": 38.6,
      "date": "2025-10-29",
      "source": {
        "url": "https://arxiv.org/abs/2510.25726",
        "kind": "paper",
        "accessed": "2026-09-04"
      },
      "conditions": {
        "split": "original",
        "notes": "Best model in the paper abstract (20.2 tool-calling turns on average); original October 2025 task release."
      }
    },
    {
      "system": "DeepSeek-V3.2-Exp",
      "developer": "DeepSeek",
      "value": 20.1,
      "date": "2025-10-29",
      "source": {
        "url": "https://arxiv.org/abs/2510.25726",
        "kind": "paper",
        "accessed": "2026-09-04"
      },
      "conditions": {
        "split": "original",
        "notes": "Top open-weights model in the paper abstract; original October 2025 task release."
      }
    },
    {
      "system": "Claude Opus 4.8 (max)",
      "developer": "Anthropic",
      "value": 76.2,
      "date": "2026-06-30",
      "source": {
        "url": "https://toolathlon.xyz/docs/leaderboard",
        "kind": "official-leaderboard",
        "accessed": "2026-09-04"
      },
      "conditions": {
        "split": "verified",
        "scaffold": "Default",
        "reasoning_effort": "max",
        "notes": "Evaluated by the Toolathlon team; +/-3.4 over 3 runs; Pass@3 84.3, Pass^3 66.7; 19.9 turns, 36.3 tool calls per task."
      }
    },
    {
      "system": "GPT-5.5 (xhigh)",
      "developer": "OpenAI",
      "value": 73.5,
      "date": "2026-06-30",
      "source": {
        "url": "https://toolathlon.xyz/docs/leaderboard",
        "kind": "official-leaderboard",
        "accessed": "2026-09-04"
      },
      "conditions": {
        "split": "verified",
        "scaffold": "Default",
        "reasoning_effort": "xhigh",
        "notes": "Evaluated by the Toolathlon team; +/-1.2 over 3 runs; Pass@3 82.4, Pass^3 62.0."
      }
    },
    {
      "system": "Kimi K3 (max)",
      "developer": "Moonshot AI",
      "value": 76.5,
      "date": "2026-07-16",
      "source": {
        "url": "https://toolathlon.xyz/docs/leaderboard",
        "kind": "official-leaderboard",
        "accessed": "2026-09-04"
      },
      "conditions": {
        "split": "verified",
        "scaffold": "Default",
        "reasoning_effort": "max",
        "notes": "Evaluated by the Toolathlon team; +/-1.9 over 3 runs; Pass@3 83.3, Pass^3 68.5 (highest Pass^3 on the board)."
      }
    },
    {
      "system": "GLM 5.3 Flash (max)",
      "developer": "Z.ai",
      "value": 78.4,
      "date": "2026-08-30",
      "source": {
        "url": "https://toolathlon.xyz/docs/leaderboard",
        "kind": "official-leaderboard",
        "accessed": "2026-09-04"
      },
      "conditions": {
        "split": "verified",
        "scaffold": "Default",
        "reasoning_effort": "max",
        "notes": "Rank 1 Pass@1 as of access date; +/-1.9 over 3 runs; Pass@3 / Pass^3 not yet published."
      }
    }
  ]
}