{
  "api_version": "1",
  "updated": "2026-10-26",
  "latest_event": "2026-10-24",
  "rights": "© Frontier Log, all rights reserved.",
  "model": {
    "slug": "claude-sonnet-4",
    "name": "Claude Sonnet 4",
    "lab": "anthropic",
    "date": "2025-05-22",
    "source": "https://www.anthropic.com/news/claude-4",
    "source_title": "Introducing Claude 4",
    "flagship": false,
    "event": "https://frontierlog.ai/event/2025-05-22-claude-4",
    "releases": [],
    "variant_of": null,
    "made_by": "Anthropic",
    "url": "https://frontierlog.ai/model/claude-sonnet-4",
    "results": [
      {
        "id": "claude-sonnet-4:swe-bench-verified:default:1",
        "model": "claude-sonnet-4",
        "model_name": "Claude Sonnet 4",
        "model_date": "2025-05-22",
        "event": "https://frontierlog.ai/event/2025-05-22-claude-4",
        "benchmark": "swe-bench-verified",
        "benchmark_name": "SWE-bench Verified",
        "setting": "default",
        "setting_name": "",
        "score": 72.7,
        "unit": "%",
        "printed": "72.7%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 4",
        "source": "https://www.anthropic.com/news/claude-4",
        "source_title": "Introducing Claude 4",
        "source_date": "2025-05-22",
        "where": "'Software engineering, SWE-bench verified' bar chart (image), bar 'Sonnet 4', value inside the bar",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "no extended thinking; bash and file-editing tools; full 500 problems",
        "caveat": "",
        "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/news/claude-opus-4-1",
            "name": "Claude Opus 4.1",
            "title": "Claude Opus 4.1",
            "lab": "anthropic",
            "date": "2025-08-05",
            "printed": "72.7%",
            "where": "Benchmark table image, row 'Agentic coding SWE-bench Verified', column 'Claude Sonnet 4'"
          },
          {
            "url": "https://www.anthropic.com/news/claude-sonnet-4-5",
            "name": "Claude Agent SDK",
            "title": "Introducing Claude Sonnet 4.5",
            "lab": "anthropic",
            "date": "2025-09-29",
            "printed": "72.7%",
            "where": "Benchmark table image, row 'Agentic coding SWE-bench Verified', column 'Claude Sonnet 4', upper value"
          },
          {
            "url": "https://www.kimi.ai/blog/kimi-k2",
            "name": "Kimi K2",
            "title": "Kimi K2: Open Agentic Intelligence",
            "lab": "moonshot",
            "date": "2025-07-11",
            "printed": "72.7",
            "where": "Evaluation Results table (Kimi-K2-Instruct tab), row SWE-bench Verified (Agentic Coding) Single Attempt (Acc) ; column Claude Sonnet 4 (w/o extended thinking)"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/5ed59708a2aeef89.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/09a6d5aa47c25cb2037efff9f486da4918f77708-3840x2304.png"
      },
      {
        "id": "claude-sonnet-4:swe-bench-verified:default:2",
        "model": "claude-sonnet-4",
        "model_name": "Claude Sonnet 4",
        "model_date": "2025-05-22",
        "event": "https://frontierlog.ai/event/2025-05-22-claude-4",
        "benchmark": "swe-bench-verified",
        "benchmark_name": "SWE-bench Verified",
        "setting": "default",
        "setting_name": "",
        "score": 80.2,
        "unit": "%",
        "printed": "80.2%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 4",
        "source": "https://www.anthropic.com/news/claude-4",
        "source_title": "Introducing Claude 4",
        "source_date": "2025-05-22",
        "where": "'Software engineering, SWE-bench verified' bar chart (image), bar 'Sonnet 4', label above the bar",
        "own": true,
        "primary": false,
        "parallel": true,
        "notes": "parallel test-time compute",
        "caveat": "",
        "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/news/claude-sonnet-4-5",
            "name": "Claude Agent SDK",
            "title": "Introducing Claude Sonnet 4.5",
            "lab": "anthropic",
            "date": "2025-09-29",
            "printed": "80.2%",
            "where": "Benchmark table image, row 'Agentic coding SWE-bench Verified', column 'Claude Sonnet 4', lower value 'with parallel test-time compute'"
          },
          {
            "url": "https://www.kimi.ai/blog/kimi-k2",
            "name": "Kimi K2",
            "title": "Kimi K2: Open Agentic Intelligence",
            "lab": "moonshot",
            "date": "2025-07-11",
            "printed": "80.2",
            "where": "Evaluation Results table (Kimi-K2-Instruct tab), row SWE-bench Verified (Agentic Coding) Multiple Attempts (Acc) ; column Claude Sonnet 4 (w/o extended thinking)"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/5ed59708a2aeef89.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/09a6d5aa47c25cb2037efff9f486da4918f77708-3840x2304.png"
      },
      {
        "id": "claude-sonnet-4:gpqa-diamond:default:1",
        "model": "claude-sonnet-4",
        "model_name": "Claude Sonnet 4",
        "model_date": "2025-05-22",
        "event": "https://frontierlog.ai/event/2025-05-22-claude-4",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 70.0,
        "unit": "%",
        "printed": "70.0%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 4",
        "source": "https://www.anthropic.com/news/claude-4",
        "source_title": "Introducing Claude 4",
        "source_date": "2025-05-22",
        "where": "Appendix, 'Performance benchmark reporting', extended-thinking list, GPQA Diamond item",
        "own": true,
        "primary": false,
        "parallel": false,
        "notes": "no extended thinking",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-17",
        "check": "human",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://www.kimi.ai/blog/kimi-k2",
            "name": "Kimi K2",
            "title": "Kimi K2: Open Agentic Intelligence",
            "lab": "moonshot",
            "date": "2025-07-11",
            "printed": "70.0",
            "where": "Evaluation Results table (Kimi-K2-Instruct tab), row GPQA-Diamond Avg@8 ; column Claude Sonnet 4 (w/o extended thinking)"
          }
        ]
      },
      {
        "id": "claude-sonnet-4:gpqa-diamond:default:2",
        "model": "claude-sonnet-4",
        "model_name": "Claude Sonnet 4",
        "model_date": "2025-05-22",
        "event": "https://frontierlog.ai/event/2025-05-22-claude-4",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 75.4,
        "unit": "%",
        "printed": "75.4%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude Sonnet 4",
        "source": "https://www.anthropic.com/news/claude-4",
        "source_title": "Introducing Claude 4",
        "source_date": "2025-05-22",
        "where": "benchmark table drawn by the page's script (not in the HTML source; render the page), row 'Graduate-level reasoning, GPQA Diamond', column 'Claude Sonnet 4', first value of the cell",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "extended thinking up to 64K tokens",
        "caveat": "",
        "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/news/claude-opus-4-1",
            "name": "Claude Opus 4.1",
            "title": "Claude Opus 4.1",
            "lab": "anthropic",
            "date": "2025-08-05",
            "printed": "75.4%",
            "where": "Benchmark table image, row 'Graduate-level reasoning GPQA Diamond', column 'Claude Sonnet 4'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/4e6f6af599737101.png"
      },
      {
        "id": "claude-sonnet-4:gpqa-diamond:default:3",
        "model": "claude-sonnet-4",
        "model_name": "Claude Sonnet 4",
        "model_date": "2025-05-22",
        "event": "https://frontierlog.ai/event/2025-05-22-claude-4",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 83.8,
        "unit": "%",
        "printed": "83.8%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude Sonnet 4",
        "source": "https://www.anthropic.com/news/claude-4",
        "source_title": "Introducing Claude 4",
        "source_date": "2025-05-22",
        "where": "benchmark table drawn by the page's script (not in the HTML source; render the page), row 'Graduate-level reasoning, GPQA Diamond', column 'Claude Sonnet 4', second value of the cell",
        "own": true,
        "primary": false,
        "parallel": true,
        "notes": "parallel test-time compute: best of several samples by an internal scoring model",
        "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/4e6f6af599737101.png"
      }
    ]
  }
}