{
  "api_version": "1",
  "updated": "2026-10-26",
  "latest_event": "2026-10-24",
  "rights": "© Frontier Log, all rights reserved.",
  "model": {
    "slug": "claude-sonnet-5",
    "name": "Claude Sonnet 5",
    "lab": "anthropic",
    "date": "2026-06-30",
    "source": "https://www.anthropic.com/news/claude-sonnet-5",
    "source_title": "Introducing Claude Sonnet 5",
    "flagship": false,
    "event": "https://frontierlog.ai/event/2026-06-30-claude-sonnet-5",
    "releases": [],
    "variant_of": null,
    "made_by": "Anthropic",
    "url": "https://frontierlog.ai/model/claude-sonnet-5",
    "results": [
      {
        "id": "claude-sonnet-5:hle:no-tools:1",
        "model": "claude-sonnet-5",
        "model_name": "Claude Sonnet 5",
        "model_date": "2026-06-30",
        "event": "https://frontierlog.ai/event/2026-06-30-claude-sonnet-5",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 43.2,
        "unit": "%",
        "printed": "43.2%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 5",
        "source": "https://www.anthropic.com/news/claude-sonnet-5",
        "source_title": "Introducing Claude Sonnet 5",
        "source_date": "2026-06-30",
        "where": "benchmark table (image), row 'Multidisciplinary reasoning, Humanity's Last Exam', column 'Sonnet 5', 'no tools'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-16",
        "check": "human",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://www.anthropic.com/claude-sonnet-5-system-card",
            "name": "Claude Sonnet 5 card",
            "title": "Claude Sonnet 5 System Card",
            "lab": "anthropic",
            "date": "2026-06-30",
            "printed": "43.2",
            "where": "Table 8.1.A (p.115), row Humanity's Last Exam / No tools, column Claude Sonnet 5"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/849d6a908f0d46ac.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/9941d610909f28a504e16dd5af823df172ec6035-2600x1234.png"
      },
      {
        "id": "claude-sonnet-5:hle:no-tools:2",
        "model": "claude-sonnet-5",
        "model_name": "Claude Sonnet 5",
        "model_date": "2026-06-30",
        "event": "https://frontierlog.ai/event/2026-06-30-claude-sonnet-5",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 43.1,
        "unit": "%",
        "printed": "43.1",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude Sonnet 5",
        "source": "https://www.anthropic.com/claude-sonnet-5-5-system-card",
        "source_title": "Claude Sonnet 5.5 System Card",
        "source_date": "2026-09-28",
        "where": "Section 8.1 evaluation summary table, row 'Humanity's Last Exam, No tools', column 'Claude Sonnet 5'",
        "own": false,
        "primary": false,
        "parallel": false,
        "notes": "",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-16",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/53b68efd4b35b2e3.png"
      },
      {
        "id": "claude-sonnet-5:browsecomp:default:1",
        "model": "claude-sonnet-5",
        "model_name": "Claude Sonnet 5",
        "model_date": "2026-06-30",
        "event": "https://frontierlog.ai/event/2026-06-30-claude-sonnet-5",
        "benchmark": "browsecomp",
        "benchmark_name": "BrowseComp",
        "setting": "default",
        "setting_name": "",
        "score": 84.7,
        "unit": "%",
        "printed": "84.7",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude Sonnet 5",
        "source": "https://www.anthropic.com/claude-sonnet-5-system-card",
        "source_title": "Claude Sonnet 5 System Card",
        "source_date": "2026-06-30",
        "where": "Section 8.1 evaluation summary table (page 115), row 'BrowseComp', column 'Claude Sonnet 5', the '(single agent)' value",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "adaptive thinking, max effort; 10M-token limit, compaction at 200K tokens",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-17",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/83869c127dd1cbc7.png"
      },
      {
        "id": "claude-sonnet-5:swe-bench-pro:default:1",
        "model": "claude-sonnet-5",
        "model_name": "Claude Sonnet 5",
        "model_date": "2026-06-30",
        "event": "https://frontierlog.ai/event/2026-06-30-claude-sonnet-5",
        "benchmark": "swe-bench-pro",
        "benchmark_name": "SWE-bench Pro",
        "setting": "default",
        "setting_name": "",
        "score": 63.2,
        "unit": "%",
        "printed": "63.2%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 5",
        "source": "https://www.anthropic.com/news/claude-sonnet-5",
        "source_title": "Introducing Claude Sonnet 5",
        "source_date": "2026-06-30",
        "where": "benchmark table (image), row 'Agentic coding, SWE-bench Pro', column 'Sonnet 5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "",
        "caveat": "The source doesn't say whether this is the public or the commercial set",
        "human_verified": true,
        "human_verified_on": "2026-10-17",
        "check": "human",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://www.anthropic.com/claude-sonnet-5-5-system-card",
            "name": "Claude Sonnet 5.5 card",
            "title": "Claude Sonnet 5.5 System Card",
            "lab": "anthropic",
            "date": "2026-09-28",
            "printed": "63.2",
            "where": "Table 8.1.A (p.109), row SWE-Bench Pro, column Claude Sonnet 5"
          },
          {
            "url": "https://www.anthropic.com/claude-sonnet-5-system-card",
            "name": "Claude Sonnet 5 card",
            "title": "Claude Sonnet 5 System Card",
            "lab": "anthropic",
            "date": "2026-06-30",
            "printed": "63.2",
            "where": "Table 8.1.A (p.115), row SWE-bench Pro, column Claude Sonnet 5"
          },
          {
            "url": "https://deepmind.google/models/model-cards/gemini-3-6-flash/",
            "name": "Gemini 3.6 Flash and 3.5 Flash-Lite card",
            "title": "Gemini 3.6 Flash - Model Card",
            "lab": "google",
            "date": "2026-07-21",
            "printed": "63.2%",
            "where": "HTML results table 'Results as of July, 2026'; row 'SWE-Bench Pro (Public)'; column claude-sonnet-5"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/849d6a908f0d46ac.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/9941d610909f28a504e16dd5af823df172ec6035-2600x1234.png"
      },
      {
        "id": "claude-sonnet-5:hle:with-tools:1",
        "model": "claude-sonnet-5",
        "model_name": "Claude Sonnet 5",
        "model_date": "2026-06-30",
        "event": "https://frontierlog.ai/event/2026-06-30-claude-sonnet-5",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "with-tools",
        "setting_name": "With tools",
        "score": 57.4,
        "unit": "%",
        "printed": "57.4%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 5",
        "source": "https://www.anthropic.com/news/claude-sonnet-5",
        "source_title": "Introducing Claude Sonnet 5",
        "source_date": "2026-06-30",
        "where": "benchmark table (image), row 'Multidisciplinary reasoning, Humanity's Last Exam', column 'Sonnet 5', 'with tools'",
        "own": true,
        "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-sonnet-5-system-card",
            "name": "Claude Sonnet 5 card",
            "title": "Claude Sonnet 5 System Card",
            "lab": "anthropic",
            "date": "2026-06-30",
            "printed": "57.4",
            "where": "Table 8.1.A, Humanity's Last Exam With tools row, Sonnet 5 column"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/849d6a908f0d46ac.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/9941d610909f28a504e16dd5af823df172ec6035-2600x1234.png"
      },
      {
        "id": "claude-sonnet-5:hle:with-tools:2",
        "model": "claude-sonnet-5",
        "model_name": "Claude Sonnet 5",
        "model_date": "2026-06-30",
        "event": "https://frontierlog.ai/event/2026-06-30-claude-sonnet-5",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "with-tools",
        "setting_name": "With tools",
        "score": 54.9,
        "unit": "%",
        "printed": "54.9%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 5",
        "source": "https://www.anthropic.com/claude-sonnet-5-5",
        "source_title": "Introducing Claude Sonnet 5.5",
        "source_date": "2026-09-28",
        "where": "benchmark table (HTML), row 'Multidisciplinary reasoning, Humanity's Last Exam', column 'Sonnet 5', 'with tools'",
        "own": false,
        "primary": false,
        "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-sonnet-5-5-system-card",
            "name": "Claude Sonnet 5.5 card",
            "title": "Claude Sonnet 5.5 System Card",
            "lab": "anthropic",
            "date": "2026-09-28",
            "printed": "54.9",
            "where": "Table 8.1.A, Humanity's Last Exam With tools row, Claude Sonnet 5 column"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/0a2dc9377a66cc01.png"
      },
      {
        "id": "claude-sonnet-5:automationbench:default:1",
        "model": "claude-sonnet-5",
        "model_name": "Claude Sonnet 5",
        "model_date": "2026-06-30",
        "event": "https://frontierlog.ai/event/2026-06-30-claude-sonnet-5",
        "benchmark": "automationbench",
        "benchmark_name": "AutomationBench",
        "setting": "default",
        "setting_name": "",
        "score": 13.5,
        "unit": "%",
        "printed": "13.5",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude Sonnet 5",
        "source": "https://www.anthropic.com/claude-sonnet-5-system-card",
        "source_title": "Claude Sonnet 5 System Card",
        "source_date": "2026-06-30",
        "where": "Section 8.1, Table 8.1.A, row 'AutomationBench', column 'Claude Sonnet 5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "Zapier leaderboard, private held-out set; max effort",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-20",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/83869c127dd1cbc7.png"
      },
      {
        "id": "claude-sonnet-5:automationbench:default:2",
        "model": "claude-sonnet-5",
        "model_name": "Claude Sonnet 5",
        "model_date": "2026-06-30",
        "event": "https://frontierlog.ai/event/2026-06-30-claude-sonnet-5",
        "benchmark": "automationbench",
        "benchmark_name": "AutomationBench",
        "setting": "default",
        "setting_name": "",
        "score": 10.7,
        "unit": "%",
        "printed": "10.7",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude Sonnet 5",
        "source": "https://www.anthropic.com/claude-sonnet-5-5-system-card",
        "source_title": "Claude Sonnet 5.5 System Card",
        "source_date": "2026-09-28",
        "where": "Section 8.1, Table 8.1.A, row 'AutomationBench', column 'Claude Sonnet 5'",
        "own": false,
        "primary": false,
        "parallel": false,
        "notes": "run by Zapier; private set, release 1.0.6",
        "caveat": "",
        "human_verified": false,
        "human_verified_on": null,
        "check": "corroborated",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://deepmind.google/models/model-cards/gemini-3-7-flash/",
            "name": "Gemini 3.7 Flash card",
            "title": "Gemini 3.7 Flash model card",
            "lab": "google",
            "date": "2026-08-13",
            "printed": "10.7%",
            "where": "Results table, row AutomationBench (Private set), column Claude Sonnet 5"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/53b68efd4b35b2e3.png"
      },
      {
        "id": "claude-sonnet-5:osworld-2-1:partial:1",
        "model": "claude-sonnet-5",
        "model_name": "Claude Sonnet 5",
        "model_date": "2026-06-30",
        "event": "https://frontierlog.ai/event/2026-06-30-claude-sonnet-5",
        "benchmark": "osworld-2-1",
        "benchmark_name": "OSWorld 2.1",
        "setting": "partial",
        "setting_name": "Full set, partial score",
        "score": 57.0,
        "unit": "%",
        "printed": "57.0%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 5",
        "source": "https://www.anthropic.com/claude-sonnet-5-5",
        "source_title": "Introducing Claude Sonnet 5.5",
        "source_date": "2026-09-28",
        "where": "benchmark table (HTML), row 'Computer use / OSWorld 2.1', column 'Sonnet 5', the value tagged 'partial'",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "max effort; mean of 5 runs; September 10, 2026 task files",
        "caveat": "",
        "human_verified": false,
        "human_verified_on": null,
        "check": "corroborated",
        "corroborated": true,
        "corroborated_on": "2026-10-20",
        "corroborated_by": [
          {
            "url": "https://www.anthropic.com/claude-sonnet-5-5-system-card",
            "name": "Claude Sonnet 5.5 card",
            "title": "Claude Sonnet 5.5 System Card",
            "lab": "anthropic",
            "date": "2026-09-28",
            "printed": "57.0%",
            "where": "section 8.13.3 OSWorld 2.1 (p. 130), sentence: 'Claude Sonnet 5 scored 57.0% partial and 25.6% strict pass'; also 8.1 Evaluation summary, row 'OSWorld 2.1', column Claude Sonnet 5: '57.0'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/0a2dc9377a66cc01.png"
      },
      {
        "id": "claude-sonnet-5:osworld-2-1:strict:1",
        "model": "claude-sonnet-5",
        "model_name": "Claude Sonnet 5",
        "model_date": "2026-06-30",
        "event": "https://frontierlog.ai/event/2026-06-30-claude-sonnet-5",
        "benchmark": "osworld-2-1",
        "benchmark_name": "OSWorld 2.1",
        "setting": "strict",
        "setting_name": "Full set, strict pass rate",
        "score": 25.6,
        "unit": "%",
        "printed": "25.6%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude Sonnet 5",
        "source": "https://www.anthropic.com/claude-sonnet-5-5-system-card",
        "source_title": "Claude Sonnet 5.5 System Card",
        "source_date": "2026-09-28",
        "where": "section 8.13.3 OSWorld 2.1, text (PDF page 130), the Claude Sonnet 5 strict pass rate",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "max effort; mean of 5 runs; September 10, 2026 task files",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-20",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/ed9b8a21a7f60448.png"
      }
    ]
  }
}