{
  "api_version": "1",
  "updated": "2026-10-26",
  "latest_event": "2026-10-24",
  "rights": "© Frontier Log, all rights reserved.",
  "model": {
    "slug": "claude-sonnet-4-5",
    "name": "Claude Sonnet 4.5",
    "lab": "anthropic",
    "date": "2025-09-29",
    "source": "https://www.anthropic.com/news/claude-sonnet-4-5",
    "source_title": "Introducing Claude Sonnet 4.5",
    "flagship": false,
    "event": "https://frontierlog.ai/event/2025-09-29-claude-sonnet-4-5",
    "releases": [],
    "variant_of": null,
    "made_by": "Anthropic",
    "url": "https://frontierlog.ai/model/claude-sonnet-4-5",
    "results": [
      {
        "id": "claude-sonnet-4-5:swe-bench-verified:default:1",
        "model": "claude-sonnet-4-5",
        "model_name": "Claude Sonnet 4.5",
        "model_date": "2025-09-29",
        "event": "https://frontierlog.ai/event/2025-09-29-claude-sonnet-4-5",
        "benchmark": "swe-bench-verified",
        "benchmark_name": "SWE-bench Verified",
        "setting": "default",
        "setting_name": "",
        "score": 77.2,
        "unit": "%",
        "printed": "77.2%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude Sonnet 4.5",
        "source": "https://www.anthropic.com/news/claude-sonnet-4-5",
        "source_title": "Introducing Claude Sonnet 4.5",
        "source_date": "2025-09-29",
        "where": "benchmark table (image), row 'Agentic coding, SWE-bench Verified', column 'Claude Sonnet 4.5', first value",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "bash and file-editing tools; 200K thinking budget; averaged over 10 trials; no test-time compute",
        "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-opus-4-5-system-card",
            "name": "Claude Opus 4.5 card",
            "title": "Claude Opus 4.5 System Card",
            "lab": "anthropic",
            "date": "2025-11-24",
            "printed": "77.2%",
            "where": "Table 2.3.A (p.19), row SWE-bench Verified, column Claude Sonnet 4.5"
          },
          {
            "url": "https://www.anthropic.com/claude-opus-4-6-system-card",
            "name": "Claude Opus 4.6 card",
            "title": "Claude Opus 4.6 System Card",
            "lab": "anthropic",
            "date": "2026-02-05",
            "printed": "77.2%",
            "where": "Table 2.3.A (p.18), row SWE-bench Verified, column Claude Sonnet 4.5"
          },
          {
            "url": "https://www.anthropic.com/news/claude-opus-4-5",
            "name": "Claude Opus 4.5",
            "title": "Introducing Claude Opus 4.5",
            "lab": "anthropic",
            "date": "2025-11-24",
            "printed": "77.2%",
            "where": "Benchmark table image, row 'Agentic coding SWE-bench Verified', column 'Sonnet 4.5'"
          },
          {
            "url": "https://www.anthropic.com/news/claude-opus-4-6",
            "name": "Claude Opus 4.6",
            "title": "Claude Opus 4.6",
            "lab": "anthropic",
            "date": "2026-02-05",
            "printed": "77.2%",
            "where": "Benchmark table image, row 'Agentic coding SWE-bench Verified', column 'Sonnet 4.5'"
          },
          {
            "url": "https://api-docs.deepseek.com/news/news251201/",
            "name": "DeepSeek-V3.2",
            "title": "DeepSeek-V3.2 Release",
            "lab": "deepseek",
            "date": "2025-12-01",
            "printed": "77.2",
            "where": "Bar chart image; group 'SWE Verified (Resolved)'; value label on Claude-4.5-Sonnet bar"
          },
          {
            "url": "https://www.minimax.io/blog/minimax-m2-en-1748600000",
            "name": "MiniMax M2",
            "title": "MiniMax M2 & Agent: Ingenious in Simplicity",
            "lab": "minimax",
            "date": "2025-10-27",
            "printed": "77.2",
            "where": "Bar chart image (8 panels); panel 'SWE-bench Verified'; value label on Claude Sonnet 4.5 bar (legend order)"
          },
          {
            "url": "https://www.minimax.io/news/minimax-m21",
            "name": "MiniMax M2.1",
            "title": "MiniMax M2.1: Significantly Enhanced Multi-Language Programming, Built for Real-World Complex Tasks",
            "lab": "minimax",
            "date": "2025-12-23",
            "printed": "77.2",
            "where": "Coding bar chart image (10 panels); panel 'SWE-bench Verified'; value label on Claude Sonnet 4.5 bar"
          },
          {
            "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": "77.2%",
            "where": "'SWE-bench Verified Score Evolution' line chart image; point label 'Sonnet 4.5 77.2%'"
          },
          {
            "url": "https://blog.google/innovation-and-ai/technology/developers-tools/gemini-3-developers/",
            "name": "Google Antigravity",
            "title": "Start building with Gemini 3",
            "lab": "google",
            "date": "2025-11-18",
            "printed": "77.2%",
            "where": "Benchmark table image; row 'SWE-Bench Verified, Single attempt'; column claude-sonnet-4-5"
          },
          {
            "url": "https://www.kimi.ai/blog/kimi-k2-thinking",
            "name": "Kimi K2 Thinking",
            "title": "Introducing Kimi K2 Thinking",
            "lab": "moonshot",
            "date": "2025-11-06",
            "printed": "77.2",
            "where": "Full Evaluations table, row SWE-bench Verified, w/ tools ; column Claude Sonnet 4.5 (Thinking)"
          },
          {
            "url": "https://mistral.ai/news/devstral-2-vibe-cli/",
            "name": "Devstral 2",
            "title": "Introducing: Devstral 2 and Mistral Vibe CLI.",
            "lab": "mistral",
            "date": "2025-12-09",
            "printed": "77.2",
            "where": "Chart 'SWE-Bench Verified: Open-weight vs Proprietary models', bar 'Claude 4.5 Sonnet'"
          },
          {
            "url": "https://mistral.ai/news/vibe-remote-agents-mistral-medium-3-5/",
            "name": "Mistral Medium 3.5",
            "title": "Remote agents in Vibe. Powered by Mistral Medium 3.5.",
            "lab": "mistral",
            "date": "2026-05-22",
            "printed": "77.2",
            "where": "Chart 'Agentic Benchmarks vs competing models', group 'SWE-Bench verified***' (footnote: self-reported; max reasoning), bar 'Claude Sonnet 4.5'"
          },
          {
            "url": "https://z.ai/blog/glm-4.6",
            "name": "GLM-4.6",
            "title": "GLM-4.6: Advanced Agentic, Reasoning and Coding Capabilities",
            "lab": "zai",
            "date": "2025-09-30",
            "printed": "77.2",
            "where": "Chart 'LLM Performance Evaluation: Agentic, Reasoning and Coding', panel 'SWE-bench Verified', bar 'Claude Sonnet 4.5'"
          },
          {
            "url": "https://z.ai/blog/glm-4.7",
            "name": "GLM-4.7",
            "title": "GLM-4.7: Advancing the Coding Capability",
            "lab": "zai",
            "date": "2025-12-22",
            "printed": "77.2",
            "where": "Benchmark table, row 'SWE-bench Verified', column 'Claude Sonnet 4.5'"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/b6f039f2d22e323a.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/67081be1ea2752e2a554e49a6aab2731b265d11b-2600x2288.png"
      },
      {
        "id": "claude-sonnet-4-5:swe-bench-verified:default:2",
        "model": "claude-sonnet-4-5",
        "model_name": "Claude Sonnet 4.5",
        "model_date": "2025-09-29",
        "event": "https://frontierlog.ai/event/2025-09-29-claude-sonnet-4-5",
        "benchmark": "swe-bench-verified",
        "benchmark_name": "SWE-bench Verified",
        "setting": "default",
        "setting_name": "",
        "score": 82.0,
        "unit": "%",
        "printed": "82.0%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude Sonnet 4.5",
        "source": "https://www.anthropic.com/news/claude-sonnet-4-5",
        "source_title": "Introducing Claude Sonnet 4.5",
        "source_date": "2025-09-29",
        "where": "benchmark table (image), row 'Agentic coding, SWE-bench Verified', column 'Claude Sonnet 4.5', second value",
        "own": true,
        "primary": false,
        "parallel": true,
        "notes": "parallel test-time compute",
        "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/b6f039f2d22e323a.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/67081be1ea2752e2a554e49a6aab2731b265d11b-2600x2288.png"
      },
      {
        "id": "claude-sonnet-4-5:gpqa-diamond:default:1",
        "model": "claude-sonnet-4-5",
        "model_name": "Claude Sonnet 4.5",
        "model_date": "2025-09-29",
        "event": "https://frontierlog.ai/event/2025-09-29-claude-sonnet-4-5",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 83.4,
        "unit": "%",
        "printed": "83.4%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude Sonnet 4.5",
        "source": "https://www.anthropic.com/news/claude-sonnet-4-5",
        "source_title": "Introducing Claude Sonnet 4.5",
        "source_date": "2025-09-29",
        "where": "benchmark table (image), row 'Graduate-level reasoning, GPQA Diamond', column 'Claude Sonnet 4.5'",
        "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-opus-4-5-system-card",
            "name": "Claude Opus 4.5 card",
            "title": "Claude Opus 4.5 System Card",
            "lab": "anthropic",
            "date": "2025-11-24",
            "printed": "83.40%",
            "where": "Table 2.3.A (p.19), row GPQA Diamond, column Claude Sonnet 4.5"
          },
          {
            "url": "https://www.anthropic.com/claude-opus-4-6-system-card",
            "name": "Claude Opus 4.6 card",
            "title": "Claude Opus 4.6 System Card",
            "lab": "anthropic",
            "date": "2026-02-05",
            "printed": "83.40%",
            "where": "Table 2.3.A (p.18), row GPQA Diamond, column Claude Sonnet 4.5"
          },
          {
            "url": "https://www.anthropic.com/news/claude-opus-4-5",
            "name": "Claude Opus 4.5",
            "title": "Introducing Claude Opus 4.5",
            "lab": "anthropic",
            "date": "2025-11-24",
            "printed": "83.4%",
            "where": "Benchmark table image, row 'Graduate-level reasoning GPQA Diamond', column 'Sonnet 4.5'"
          },
          {
            "url": "https://www.anthropic.com/news/claude-opus-4-6",
            "name": "Claude Opus 4.6",
            "title": "Claude Opus 4.6",
            "lab": "anthropic",
            "date": "2026-02-05",
            "printed": "83.4%",
            "where": "Benchmark table image, row 'Graduate-level reasoning GPQA Diamond', column 'Sonnet 4.5'"
          },
          {
            "url": "https://blog.google/innovation-and-ai/technology/developers-tools/gemini-3-developers/",
            "name": "Google Antigravity",
            "title": "Start building with Gemini 3",
            "lab": "google",
            "date": "2025-11-18",
            "printed": "83.4%",
            "where": "Benchmark table image; row 'GPQA Diamond, No tools'; column claude-sonnet-4-5"
          },
          {
            "url": "https://www.kimi.ai/blog/kimi-k2-thinking",
            "name": "Kimi K2 Thinking",
            "title": "Introducing Kimi K2 Thinking",
            "lab": "moonshot",
            "date": "2025-11-06",
            "printed": "83.4",
            "where": "Full Evaluations table, row GPQA-Diamond, no tools ; column Claude Sonnet 4.5 (Thinking)"
          },
          {
            "url": "https://z.ai/blog/glm-4.7",
            "name": "GLM-4.7",
            "title": "GLM-4.7: Advancing the Coding Capability",
            "lab": "zai",
            "date": "2025-12-22",
            "printed": "83.4",
            "where": "Benchmark table, row 'GPQA-Diamond', column 'Claude Sonnet 4.5'"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/b6f039f2d22e323a.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/67081be1ea2752e2a554e49a6aab2731b265d11b-2600x2288.png"
      },
      {
        "id": "claude-sonnet-4-5:hle:no-tools:1",
        "model": "claude-sonnet-4-5",
        "model_name": "Claude Sonnet 4.5",
        "model_date": "2025-09-29",
        "event": "https://frontierlog.ai/event/2025-09-29-claude-sonnet-4-5",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 17.7,
        "unit": "%",
        "printed": "17.7%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 4.5",
        "source": "https://www.anthropic.com/claude-opus-4-5-system-card",
        "source_title": "Claude Opus 4.5 System Card",
        "source_date": "2025-11-24",
        "where": "Figure 2.16.A 'Humanity's Last Exam (HLE)', panel 'Without Search', bar 'Sonnet 4.5' (page 32)",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "decontamination improved with Opus 4.5, which may affect this score",
        "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-opus-4-6",
            "name": "Claude Opus 4.6",
            "title": "Claude Opus 4.6",
            "lab": "anthropic",
            "date": "2026-02-05",
            "printed": "17.7%",
            "where": "Benchmark table image, row 'Multidisciplinary reasoning Humanity's Last Exam, 'without tools'', column 'Sonnet 4.5'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/119c3ba76f816d86.png"
      },
      {
        "id": "claude-sonnet-4-5:arc-agi-2:default:1",
        "model": "claude-sonnet-4-5",
        "model_name": "Claude Sonnet 4.5",
        "model_date": "2025-09-29",
        "event": "https://frontierlog.ai/event/2025-09-29-claude-sonnet-4-5",
        "benchmark": "arc-agi-2",
        "benchmark_name": "ARC-AGI-2",
        "setting": "default",
        "setting_name": "",
        "score": 13.6,
        "unit": "%",
        "printed": "13.6%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 4.5",
        "source": "https://www.anthropic.com/news/claude-opus-4-5",
        "source_title": "Introducing Claude Opus 4.5",
        "source_date": "2025-11-24",
        "where": "benchmark table (image), row 'Novel problem solving, ARC-AGI-2 (Verified)', column 'Sonnet 4.5'",
        "own": false,
        "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-opus-4-5-system-card",
            "name": "Claude Opus 4.5 card",
            "title": "Claude Opus 4.5 System Card",
            "lab": "anthropic",
            "date": "2025-11-24",
            "printed": "13.6%",
            "where": "Table 2.3.A (p.19), row ARC-AGI-2 (Verified), column Claude Sonnet 4.5"
          },
          {
            "url": "https://www.anthropic.com/claude-opus-4-6-system-card",
            "name": "Claude Opus 4.6 card",
            "title": "Claude Opus 4.6 System Card",
            "lab": "anthropic",
            "date": "2026-02-05",
            "printed": "13.6%",
            "where": "Table 2.3.A (p.18), row ARC-AGI-2 (Verified), column Claude Sonnet 4.5"
          },
          {
            "url": "https://www.anthropic.com/news/claude-opus-4-6",
            "name": "Claude Opus 4.6",
            "title": "Claude Opus 4.6",
            "lab": "anthropic",
            "date": "2026-02-05",
            "printed": "13.6%",
            "where": "Benchmark table image, row 'Novel problem-solving ARC AGI 2', column 'Sonnet 4.5'"
          },
          {
            "url": "https://blog.google/innovation-and-ai/technology/developers-tools/gemini-3-developers/",
            "name": "Google Antigravity",
            "title": "Start building with Gemini 3",
            "lab": "google",
            "date": "2025-11-18",
            "printed": "13.6%",
            "where": "Benchmark table image; row 'ARC-AGI-2, ARC Prize Verified'; column claude-sonnet-4-5"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/790b0c5480216588.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/52303b11db76017fd0c2f73c7fafa5c752515979-2600x2236.png"
      },
      {
        "id": "claude-sonnet-4-5:terminal-bench-2-0:default:1",
        "model": "claude-sonnet-4-5",
        "model_name": "Claude Sonnet 4.5",
        "model_date": "2025-09-29",
        "event": "https://frontierlog.ai/event/2025-09-29-claude-sonnet-4-5",
        "benchmark": "terminal-bench-2-0",
        "benchmark_name": "Terminal-Bench 2.0",
        "setting": "default",
        "setting_name": "",
        "score": 50.0,
        "unit": "%",
        "printed": "50.0%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 4.5",
        "source": "https://www.anthropic.com/news/claude-opus-4-5",
        "source_title": "Introducing Claude Opus 4.5",
        "source_date": "2025-11-24",
        "where": "benchmark table (image), row 'Agentic terminal coding, Terminal-bench 2.0', column 'Sonnet 4.5'",
        "own": false,
        "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-opus-4-5-system-card",
            "name": "Claude Opus 4.5 card",
            "title": "Claude Opus 4.5 System Card",
            "lab": "anthropic",
            "date": "2025-11-24",
            "printed": "50.0%",
            "where": "Table 2.3.A (p.19), row Terminal-bench 2.0, column Claude Sonnet 4.5"
          },
          {
            "url": "https://www.minimax.io/news/minimax-m21",
            "name": "MiniMax M2.1",
            "title": "MiniMax M2.1: Significantly Enhanced Multi-Language Programming, Built for Real-World Complex Tasks",
            "lab": "minimax",
            "date": "2025-12-23",
            "printed": "50",
            "where": "Coding bar chart image (10 panels); panel 'Terminal-bench 2.0'; value label on Claude Sonnet 4.5 bar"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/790b0c5480216588.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/52303b11db76017fd0c2f73c7fafa5c752515979-2600x2236.png"
      },
      {
        "id": "claude-sonnet-4-5:hle:with-tools:1",
        "model": "claude-sonnet-4-5",
        "model_name": "Claude Sonnet 4.5",
        "model_date": "2025-09-29",
        "event": "https://frontierlog.ai/event/2025-09-29-claude-sonnet-4-5",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "with-tools",
        "setting_name": "With tools",
        "score": 28.4,
        "unit": "%",
        "printed": "28.4%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 4.5",
        "source": "https://www.anthropic.com/claude-opus-4-5-system-card",
        "source_title": "Claude Opus 4.5 System Card",
        "source_date": "2025-11-24",
        "where": "Figure 2.16.A 'Humanity's Last Exam (HLE)', panel 'With Search', bar 'Sonnet 4.5' (page 32)",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "web search, web fetch and code execution, no reasoning; Claude Sonnet 4.5 as grader",
        "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/119c3ba76f816d86.png"
      },
      {
        "id": "claude-sonnet-4-5:hle:with-tools:2",
        "model": "claude-sonnet-4-5",
        "model_name": "Claude Sonnet 4.5",
        "model_date": "2025-09-29",
        "event": "https://frontierlog.ai/event/2025-09-29-claude-sonnet-4-5",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "with-tools",
        "setting_name": "With tools",
        "score": 33.6,
        "unit": "%",
        "printed": "33.6%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 4.5",
        "source": "https://www.anthropic.com/news/claude-opus-4-6",
        "source_title": "Claude Opus 4.6",
        "source_date": "2026-02-05",
        "where": "benchmark table (image), row 'Multidisciplinary reasoning, Humanity's Last Exam', column 'Sonnet 4.5', 'with tools'",
        "own": false,
        "primary": false,
        "parallel": false,
        "notes": "web search, web fetch, code execution, programmatic tool calling; compaction at 50K up to 3M tokens; max effort, adaptive 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.anthropic.com/claude-opus-4-6-system-card",
            "name": "Claude Opus 4.6 card",
            "title": "Claude Opus 4.6 System Card",
            "lab": "anthropic",
            "date": "2026-02-05",
            "printed": "33.6%",
            "where": "Figure 2.21.2.A (p.42), right panel With Tools, bar Sonnet 4.5 value label"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/e6c54904fe1e4f51.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/f9564dd2f758237bd9dbe775674c4a375aff1e8a-2600x2968.png"
      },
      {
        "id": "claude-sonnet-4-5:browsecomp:default:1",
        "model": "claude-sonnet-4-5",
        "model_name": "Claude Sonnet 4.5",
        "model_date": "2025-09-29",
        "event": "https://frontierlog.ai/event/2025-09-29-claude-sonnet-4-5",
        "benchmark": "browsecomp",
        "benchmark_name": "BrowseComp",
        "setting": "default",
        "setting_name": "",
        "score": 43.9,
        "unit": "%",
        "printed": "43.9%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Sonnet 4.5",
        "source": "https://www.anthropic.com/news/claude-opus-4-6",
        "source_title": "Claude Opus 4.6",
        "source_date": "2026-02-05",
        "where": "benchmark table (image), row 'Agentic search, BrowseComp', column 'Sonnet 4.5'",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-17",
        "check": "human",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://mistral.ai/news/vibe-remote-agents-mistral-medium-3-5/",
            "name": "Mistral Medium 3.5",
            "title": "Remote agents in Vibe. Powered by Mistral Medium 3.5.",
            "lab": "mistral",
            "date": "2026-05-22",
            "printed": "43.9",
            "where": "Chart 'Agentic Benchmarks vs competing models', group 'BrowseComp****' (footnote: self-reported, Mistral uses context management, discard-all at 100k), bar 'Claude Sonnet 4.5'"
          }
        ],
        "source_image": "https://frontierlog.ai/sources/e6c54904fe1e4f51.png",
        "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/f9564dd2f758237bd9dbe775674c4a375aff1e8a-2600x2968.png"
      },
      {
        "id": "claude-sonnet-4-5:terminal-bench-2-0:default:2",
        "model": "claude-sonnet-4-5",
        "model_name": "Claude Sonnet 4.5",
        "model_date": "2025-09-29",
        "event": "https://frontierlog.ai/event/2025-09-29-claude-sonnet-4-5",
        "benchmark": "terminal-bench-2-0",
        "benchmark_name": "Terminal-Bench 2.0",
        "setting": "default",
        "setting_name": "",
        "score": 51.0,
        "unit": "%",
        "printed": "51.0%",
        "reported_by": "anthropic",
        "run_by": null,
        "label": "Claude Sonnet 4.5",
        "source": "https://www.anthropic.com/claude-opus-4-6-system-card",
        "source_title": "Claude Opus 4.6 System Card",
        "source_date": "2026-02-05",
        "where": "Table 2.3.A (PDF page 18), row 'Terminal-Bench 2.0', column 'Claude Sonnet 4.5'",
        "own": false,
        "primary": false,
        "parallel": false,
        "notes": "",
        "caveat": "",
        "human_verified": false,
        "human_verified_on": null,
        "check": "corroborated",
        "corroborated": true,
        "corroborated_on": "2026-10-20",
        "corroborated_by": [
          {
            "url": "https://www.anthropic.com/news/claude-opus-4-6",
            "name": "Claude Opus 4.6",
            "title": "Claude Opus 4.6",
            "lab": "anthropic",
            "date": "2026-02-05",
            "printed": "51.0%",
            "where": "main benchmark table image, row 'Agentic terminal coding / Terminal-Bench 2.0', column 'Sonnet 4.5'"
          },
          {
            "url": "https://www.anthropic.com/news/claude-sonnet-4-6",
            "name": "Claude Sonnet 4.6",
            "title": "Introducing Sonnet 4.6",
            "lab": "anthropic",
            "date": "2026-02-17",
            "printed": "51.0%",
            "where": "main benchmark table image, row 'Agentic terminal coding / Terminal-Bench 2.0', column 'Sonnet 4.5'"
          },
          {
            "url": "https://www.anthropic.com/claude-sonnet-4-6-system-card",
            "name": "Claude Sonnet 4.6 card",
            "title": "Claude Sonnet 4.6 System Card",
            "lab": "anthropic",
            "date": "2026-02-17",
            "printed": "51.0%",
            "where": "section 2 evaluation summary table, row 'Terminal-Bench 2.0', column 'Claude Sonnet 4.5'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/32cbf86df2814c55.png"
      }
    ]
  }
}