{
  "api_version": "1",
  "updated": "2026-10-26",
  "latest_event": "2026-10-24",
  "rights": "© Frontier Log, all rights reserved.",
  "model": {
    "slug": "claude-3-7-sonnet",
    "name": "Claude 3.7 Sonnet",
    "lab": "anthropic",
    "date": "2025-02-24",
    "source": "https://www.anthropic.com/news/claude-3-7-sonnet",
    "source_title": "Claude 3.7 Sonnet and Claude Code",
    "flagship": true,
    "event": "https://frontierlog.ai/event/2025-02-24-claude-3-7-sonnet",
    "releases": [],
    "variant_of": null,
    "made_by": "Anthropic",
    "url": "https://frontierlog.ai/model/claude-3-7-sonnet",
    "results": [
      {
        "id": "claude-3-7-sonnet:gpqa-diamond:default:1",
        "model": "claude-3-7-sonnet",
        "model_name": "Claude 3.7 Sonnet",
        "model_date": "2025-02-24",
        "event": "https://frontierlog.ai/event/2025-02-24-claude-3-7-sonnet",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 78.2,
        "unit": "%",
        "printed": "78.2%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude 3.7 Sonnet, 64K extended thinking",
        "source": "https://www.anthropic.com/news/claude-3-7-sonnet",
        "source_title": "Claude 3.7 Sonnet and Claude Code",
        "source_date": "2025-02-24",
        "where": "benchmark table (image), row 'Graduate-level reasoning, GPQA Diamond', column 'Claude 3.7 Sonnet, 64K extended thinking', first value",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "pass@1 averaged over several trials",
        "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/news/claude-4",
            "name": "Claude Code generally available",
            "title": "Introducing Claude 4",
            "lab": "anthropic",
            "date": "2025-05-22",
            "printed": "78.2%",
            "where": "Benchmarks table (HTML, rendered), row 'Graduate-level reasoning GPQA Diamond', column 'Claude Sonnet 3.7'"
          },
          {
            "url": "https://blog.google/innovation-and-ai/models-and-research/google-deepmind/gemini-model-thinking-updates-march-2025/",
            "name": "Gemini 2.5 Pro",
            "title": "Gemini 2.5: Our most intelligent AI model",
            "lab": "google",
            "date": "2025-03-25",
            "printed": "78.2%",
            "where": "Benchmark table image; row 'GPQA diamond, single attempt (pass@1)'; column claude-3-7-sonnet"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/7018392ca041e885.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/654cf6680d32858dfba9af644f8c4a5b04425af1-2600x2360.png"
      },
      {
        "id": "claude-3-7-sonnet:gpqa-diamond:default:2",
        "model": "claude-3-7-sonnet",
        "model_name": "Claude 3.7 Sonnet",
        "model_date": "2025-02-24",
        "event": "https://frontierlog.ai/event/2025-02-24-claude-3-7-sonnet",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 84.8,
        "unit": "%",
        "printed": "84.8%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude 3.7 Sonnet, 64K extended thinking",
        "source": "https://www.anthropic.com/news/claude-3-7-sonnet",
        "source_title": "Claude 3.7 Sonnet and Claude Code",
        "source_date": "2025-02-24",
        "where": "benchmark table (image), row 'Graduate-level reasoning, GPQA Diamond', column 'Claude 3.7 Sonnet, 64K extended thinking', second value",
        "own": true,
        "primary": false,
        "parallel": true,
        "notes": "parallel test-time compute with internal scoring",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-16",
        "check": "human",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://blog.google/innovation-and-ai/models-and-research/google-deepmind/gemini-model-thinking-updates-march-2025/",
            "name": "Gemini 2.5 Pro",
            "title": "Gemini 2.5: Our most intelligent AI model",
            "lab": "google",
            "date": "2025-03-25",
            "printed": "84.8%",
            "where": "Benchmark table image; row 'GPQA diamond, multiple attempts (internal scoring with parallel test time compute for Anthropic models)'; column claude-3-7-sonnet"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/7018392ca041e885.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/654cf6680d32858dfba9af644f8c4a5b04425af1-2600x2360.png"
      },
      {
        "id": "claude-3-7-sonnet:gpqa-diamond:default:3",
        "model": "claude-3-7-sonnet",
        "model_name": "Claude 3.7 Sonnet",
        "model_date": "2025-02-24",
        "event": "https://frontierlog.ai/event/2025-02-24-claude-3-7-sonnet",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 68.0,
        "unit": "%",
        "printed": "68.0%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude 3.7 Sonnet, no extended thinking",
        "source": "https://www.anthropic.com/news/claude-3-7-sonnet",
        "source_title": "Claude 3.7 Sonnet and Claude Code",
        "source_date": "2025-02-24",
        "where": "benchmark table (image), row 'Graduate-level reasoning, GPQA Diamond', column 'Claude 3.7 Sonnet, No extended thinking'",
        "own": true,
        "primary": false,
        "parallel": false,
        "notes": "no extended thinking",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-16",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_image": "https://frontierlog.ai/sources/7018392ca041e885.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/654cf6680d32858dfba9af644f8c4a5b04425af1-2600x2360.png"
      },
      {
        "id": "claude-3-7-sonnet:swe-bench-verified:default:1",
        "model": "claude-3-7-sonnet",
        "model_name": "Claude 3.7 Sonnet",
        "model_date": "2025-02-24",
        "event": "https://frontierlog.ai/event/2025-02-24-claude-3-7-sonnet",
        "benchmark": "swe-bench-verified",
        "benchmark_name": "SWE-bench Verified",
        "setting": "default",
        "setting_name": "",
        "score": 62.3,
        "unit": "%",
        "printed": "62.3%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude 3.7 Sonnet, no extended thinking",
        "source": "https://www.anthropic.com/news/claude-3-7-sonnet",
        "source_title": "Claude 3.7 Sonnet and Claude Code",
        "source_date": "2025-02-24",
        "where": "benchmark table (image), row 'Agentic coding, SWE-bench Verified', column 'Claude 3.7 Sonnet, No extended thinking', first value",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "bash and editor tools plus a thinking tool; averaged over 16 trials",
        "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/news/claude-4",
            "name": "Claude Code generally available",
            "title": "Introducing Claude 4",
            "lab": "anthropic",
            "date": "2025-05-22",
            "printed": "62.3%",
            "where": "Benchmarks table (HTML, rendered), row 'Agentic coding SWE-bench Verified', column 'Claude Sonnet 3.7', first value"
          },
          {
            "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": "62.3%",
            "where": "Bar chart 'Software engineering, SWE-bench Verified', bar 'Sonnet 3.7'"
          },
          {
            "url": "https://www.minimax.io/news/minimax-m25",
            "name": "MiniMax M2.5",
            "title": "MiniMax M2.5: Built for Real-World Productivity.",
            "lab": "minimax",
            "date": "2026-02-12",
            "printed": "62.3%",
            "where": "'SWE-bench Verified Score Evolution' line chart image; point label 'Sonnet 3.7 62.3%'"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/7018392ca041e885.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/654cf6680d32858dfba9af644f8c4a5b04425af1-2600x2360.png"
      },
      {
        "id": "claude-3-7-sonnet:swe-bench-verified:default:2",
        "model": "claude-3-7-sonnet",
        "model_name": "Claude 3.7 Sonnet",
        "model_date": "2025-02-24",
        "event": "https://frontierlog.ai/event/2025-02-24-claude-3-7-sonnet",
        "benchmark": "swe-bench-verified",
        "benchmark_name": "SWE-bench Verified",
        "setting": "default",
        "setting_name": "",
        "score": 70.3,
        "unit": "%",
        "printed": "70.3%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude 3.7 Sonnet, no extended thinking",
        "source": "https://www.anthropic.com/news/claude-3-7-sonnet",
        "source_title": "Claude 3.7 Sonnet and Claude Code",
        "source_date": "2025-02-24",
        "where": "benchmark table (image), row 'Agentic coding, SWE-bench Verified', column 'Claude 3.7 Sonnet, No extended thinking', second value",
        "own": true,
        "primary": false,
        "parallel": true,
        "notes": "internal scoring and a custom scaffold, on a reduced subset of problems",
        "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/news/claude-4",
            "name": "Claude Code generally available",
            "title": "Introducing Claude 4",
            "lab": "anthropic",
            "date": "2025-05-22",
            "printed": "70.3%",
            "where": "Benchmarks table (HTML, rendered), row 'Agentic coding SWE-bench Verified', column 'Claude Sonnet 3.7', second value (footnote 5: parallel test-time compute)"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/7018392ca041e885.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/654cf6680d32858dfba9af644f8c4a5b04425af1-2600x2360.png"
      },
      {
        "id": "claude-3-7-sonnet:aime-2024:no-tools:1",
        "model": "claude-3-7-sonnet",
        "model_name": "Claude 3.7 Sonnet",
        "model_date": "2025-02-24",
        "event": "https://frontierlog.ai/event/2025-02-24-claude-3-7-sonnet",
        "benchmark": "aime-2024",
        "benchmark_name": "AIME 2024",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 61.3,
        "unit": "%",
        "printed": "61.3%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude 3.7 Sonnet (64K extended thinking)",
        "source": "https://www.anthropic.com/news/claude-3-7-sonnet",
        "source_title": "Claude 3.7 Sonnet and Claude Code",
        "source_date": "2025-02-24",
        "where": "benchmark table image, row 'High school math competition / AIME 2024', column 'Claude 3.7 Sonnet 64K extended thinking', first number of the cell",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "64K extended thinking; pass@1 averaged over up to 16 trials",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-18",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_image": "https://frontierlog.ai/sources/7018392ca041e885.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/654cf6680d32858dfba9af644f8c4a5b04425af1-2600x2360.png"
      },
      {
        "id": "claude-3-7-sonnet:aime-2024:no-tools:2",
        "model": "claude-3-7-sonnet",
        "model_name": "Claude 3.7 Sonnet",
        "model_date": "2025-02-24",
        "event": "https://frontierlog.ai/event/2025-02-24-claude-3-7-sonnet",
        "benchmark": "aime-2024",
        "benchmark_name": "AIME 2024",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 80.0,
        "unit": "%",
        "printed": "80.0%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude 3.7 Sonnet (64K extended thinking)",
        "source": "https://www.anthropic.com/news/claude-3-7-sonnet",
        "source_title": "Claude 3.7 Sonnet and Claude Code",
        "source_date": "2025-02-24",
        "where": "benchmark table image, row 'AIME 2024', column 'Claude 3.7 Sonnet 64K extended thinking', second number of the cell (after the slash)",
        "own": true,
        "primary": false,
        "parallel": true,
        "notes": "64K extended thinking; parallel test-time compute with internal scoring",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-18",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_image": "https://frontierlog.ai/sources/7018392ca041e885.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/654cf6680d32858dfba9af644f8c4a5b04425af1-2600x2360.png"
      },
      {
        "id": "claude-3-7-sonnet:aime-2024:no-tools:3",
        "model": "claude-3-7-sonnet",
        "model_name": "Claude 3.7 Sonnet",
        "model_date": "2025-02-24",
        "event": "https://frontierlog.ai/event/2025-02-24-claude-3-7-sonnet",
        "benchmark": "aime-2024",
        "benchmark_name": "AIME 2024",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 23.3,
        "unit": "%",
        "printed": "23.3%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude 3.7 Sonnet (No extended thinking)",
        "source": "https://www.anthropic.com/news/claude-3-7-sonnet",
        "source_title": "Claude 3.7 Sonnet and Claude Code",
        "source_date": "2025-02-24",
        "where": "benchmark table image, row 'AIME 2024', column 'Claude 3.7 Sonnet No extended thinking'",
        "own": true,
        "primary": false,
        "parallel": false,
        "notes": "no extended thinking; pass@1 averaged over up to 16 trials",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-18",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_image": "https://frontierlog.ai/sources/7018392ca041e885.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/654cf6680d32858dfba9af644f8c4a5b04425af1-2600x2360.png"
      }
    ]
  }
}