{
  "api_version": "1",
  "updated": "2026-11-10",
  "latest_event": "2026-11-09",
  "rights": "© Frontier Log, all rights reserved.",
  "benchmark": {
    "slug": "osworld-2-0-2026-08-08",
    "name": "OSWorld 2.0 (v2026.08.08)",
    "family": "osworld",
    "unit": "%",
    "higher_is_better": true,
    "status": "superseded",
    "active": false,
    "superseded_by": "osworld-2-1",
    "retired": null,
    "benchmark_url": "https://github.com/xlang-ai/OSWorld-V2/tree/v2026.08.08",
    "introduced": "2026-08-08",
    "creator": "XLANG Lab",
    "introduced_source": "https://github.com/xlang-ai/OSWorld-V2",
    "settings": [
      {
        "setting": "offline-partial",
        "name": "Offline set, partial score",
        "url": "https://frontierlog.ai/benchmark/osworld-2-0-2026-08-08",
        "results": 14
      },
      {
        "setting": "anthropic-corrected-partial",
        "name": "Anthropic's corrected set, partial score",
        "url": "https://frontierlog.ai/benchmark/osworld-2-0-2026-08-08/anthropic-corrected-partial",
        "results": 2
      },
      {
        "setting": "scoring-unstated",
        "name": "Full set, scoring not stated",
        "url": "https://frontierlog.ai/benchmark/osworld-2-0-2026-08-08/scoring-unstated",
        "results": 1
      }
    ]
  },
  "total": 2,
  "results": [
    {
      "id": "claude-opus-5:osworld-2-0-2026-08-08:anthropic-corrected-partial:1",
      "model": "claude-opus-5",
      "model_name": "Claude Opus 5",
      "model_date": "2026-07-24",
      "event": "https://frontierlog.ai/event/2026-07-24-claude-opus-5",
      "benchmark": "osworld-2-0-2026-08-08",
      "benchmark_name": "OSWorld 2.0 (v2026.08.08)",
      "setting": "anthropic-corrected-partial",
      "setting_name": "Anthropic's corrected set, partial score",
      "score": 75.4,
      "unit": "%",
      "printed": "75.4%",
      "reported_by": "anthropic",
      "run_by": null,
      "label": "Opus 5",
      "source": "https://www.anthropic.com/claude-fable-and-mythos-5-1",
      "source_title": "Introducing Claude Fable 5.1 and Claude Mythos 5.1",
      "source_date": "2026-09-01",
      "where": "benchmark table (HTML), row 'Computer use / OSWorld 2.0 [2]', column 'Opus 5', the value tagged 'partial'",
      "own": false,
      "primary": true,
      "parallel": false,
      "notes": "",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-19",
      "check": "human",
      "corroborated": true,
      "corroborated_on": "2026-10-19",
      "corroborated_by": [
        {
          "url": "https://www.anthropic.com/claude-fable-5-1-mythos-5-1-system-card",
          "name": "Claude Fable 5.1 and Mythos 5.1 card",
          "title": "Claude Fable 5.1 and Claude Mythos 5.1 System Card",
          "lab": "anthropic",
          "date": "2026-09-01",
          "printed": "75.4%",
          "where": "section 8.14.3 sentence: 'Claude Opus 5 scored 75.4% partial' on the August 2026 task release (Table 8.1.A OSWorld 2.0 (partial/strict) prints 75.4/39.6)"
        }
      ],
      "source_capture": "https://frontierlog.ai/captures/f381a9ad69b68272.png"
    },
    {
      "id": "claude-fable-5-1:osworld-2-0-2026-08-08:anthropic-corrected-partial:1",
      "model": "claude-fable-5-1",
      "model_name": "Claude Fable 5.1",
      "model_date": "2026-09-01",
      "event": "https://frontierlog.ai/event/2026-09-01-claude-fable-5-1-mythos-5-1",
      "benchmark": "osworld-2-0-2026-08-08",
      "benchmark_name": "OSWorld 2.0 (v2026.08.08)",
      "setting": "anthropic-corrected-partial",
      "setting_name": "Anthropic's corrected set, partial score",
      "score": 77.9,
      "unit": "%",
      "printed": "77.9%",
      "reported_by": "anthropic",
      "run_by": null,
      "label": "Fable 5.1",
      "source": "https://www.anthropic.com/claude-fable-and-mythos-5-1",
      "source_title": "Introducing Claude Fable 5.1 and Claude Mythos 5.1",
      "source_date": "2026-09-01",
      "where": "benchmark table (HTML), row 'Computer use / OSWorld 2.0 [2]', column 'Fable 5.1', the value tagged 'partial'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "tasks where safeguards intervened scored zero",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-18",
      "check": "human",
      "corroborated": false,
      "corroborated_on": null,
      "corroborated_by": [],
      "source_capture": "https://frontierlog.ai/captures/f381a9ad69b68272.png"
    }
  ]
}