{
  "api_version": "1",
  "updated": "2026-10-16",
  "latest_event": "2026-10-15",
  "rights": "© Frontier Log, all rights reserved.",
  "benchmarks": [
    {
      "slug": "hle",
      "name": "Humanity's Last Exam",
      "family": "hle",
      "unit": "%",
      "higher_is_better": true,
      "status": "active",
      "benchmark_url": "https://lastexam.ai",
      "settings": [
        {
          "setting": "no-tools",
          "name": "No tools",
          "url": "https://frontierlog.ai/benchmark/hle",
          "results": 17
        },
        {
          "setting": "with-tools",
          "name": "With tools",
          "url": "https://frontierlog.ai/benchmark/hle/with-tools",
          "results": 2
        }
      ]
    },
    {
      "slug": "swe-bench-verified",
      "name": "SWE-bench Verified",
      "family": "swe-bench",
      "unit": "%",
      "higher_is_better": true,
      "status": "retired",
      "benchmark_url": "https://www.swebench.com",
      "settings": [
        {
          "setting": "default",
          "name": "",
          "url": "https://frontierlog.ai/benchmark/swe-bench-verified",
          "results": 11
        }
      ]
    },
    {
      "slug": "gpqa-diamond",
      "name": "GPQA Diamond",
      "family": "gpqa",
      "unit": "%",
      "higher_is_better": true,
      "status": "retired",
      "benchmark_url": "https://github.com/idavidrein/gpqa",
      "settings": [
        {
          "setting": "default",
          "name": "",
          "url": "https://frontierlog.ai/benchmark/gpqa-diamond",
          "results": 11
        }
      ]
    },
    {
      "slug": "arc-agi-2",
      "name": "ARC-AGI-2",
      "family": "arc-agi",
      "unit": "%",
      "higher_is_better": true,
      "status": "active",
      "benchmark_url": "https://arcprize.org/arc-agi/2/",
      "settings": [
        {
          "setting": "default",
          "name": "",
          "url": "https://frontierlog.ai/benchmark/arc-agi-2",
          "results": 9
        }
      ]
    },
    {
      "slug": "terminal-bench-2-0",
      "name": "Terminal-Bench 2.0",
      "family": "terminal-bench",
      "unit": "%",
      "higher_is_better": true,
      "status": "superseded",
      "benchmark_url": "https://www.tbench.ai",
      "settings": [
        {
          "setting": "default",
          "name": "",
          "url": "https://frontierlog.ai/benchmark/terminal-bench-2-0",
          "results": 9
        }
      ]
    },
    {
      "slug": "terminal-bench-4-0",
      "name": "Terminal-Bench 4.0",
      "family": "terminal-bench",
      "unit": "%",
      "higher_is_better": true,
      "status": "active",
      "benchmark_url": "https://www.tbench.ai",
      "settings": [
        {
          "setting": "default",
          "name": "",
          "url": "https://frontierlog.ai/benchmark/terminal-bench-4-0",
          "results": 3
        }
      ]
    },
    {
      "slug": "terminal-bench-science-0-1",
      "name": "Terminal-Bench Science 0.1",
      "family": "terminal-bench-science",
      "unit": "%",
      "higher_is_better": true,
      "status": "active",
      "benchmark_url": "https://www.tbench.ai",
      "settings": [
        {
          "setting": "default",
          "name": "",
          "url": "https://frontierlog.ai/benchmark/terminal-bench-science-0-1",
          "results": 3
        }
      ]
    },
    {
      "slug": "automationbench",
      "name": "AutomationBench",
      "family": "automationbench",
      "unit": "%",
      "higher_is_better": true,
      "status": "active",
      "benchmark_url": "https://zapier.com/benchmarks",
      "settings": [
        {
          "setting": "default",
          "name": "",
          "url": "https://frontierlog.ai/benchmark/automationbench",
          "results": 3
        }
      ]
    },
    {
      "slug": "osworld-2-0",
      "name": "OSWorld 2.0",
      "family": "osworld",
      "unit": "%",
      "higher_is_better": true,
      "status": "active",
      "benchmark_url": "https://huggingface.co/papers/2606.29537",
      "settings": [
        {
          "setting": "offline-partial",
          "name": "Offline set, partial score",
          "url": "https://frontierlog.ai/benchmark/osworld-2-0",
          "results": 2
        }
      ]
    }
  ]
}
