{
  "api_version": "1",
  "updated": "2026-10-26",
  "latest_event": "2026-10-24",
  "rights": "© Frontier Log, all rights reserved.",
  "model": {
    "slug": "gpt-5-5",
    "name": "GPT-5.5",
    "lab": "openai",
    "date": "2026-04-23",
    "flagship": true,
    "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
    "releases": [],
    "variant_of": null,
    "made_by": "OpenAI",
    "url": "https://frontierlog.ai/model/gpt-5-5",
    "results": [
      {
        "id": "gpt-5-5:gpqa-diamond:default:1",
        "model": "gpt-5-5",
        "model_name": "GPT-5.5",
        "model_date": "2026-04-23",
        "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 93.6,
        "unit": "%",
        "printed": "93.6%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.5",
        "source": "https://openai.com/index/introducing-gpt-5-5",
        "source_title": "Introducing GPT-5.5",
        "source_date": "2026-04-23",
        "where": "academic eval table, column 'GPT‑5.5', row 'GPQA Diamond'",
        "own": true,
        "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://openai.com/index/gpt-5-6",
            "name": "GPT-5.6",
            "title": "GPT-5.6: Frontier intelligence that scales with your ambition",
            "lab": "openai",
            "date": "2026-07-09",
            "printed": "93.6%",
            "where": "table 7 \"GPQA Diamond\" row, \"GPT-5.5\" column"
          },
          {
            "url": "https://z.ai/blog/glm-5.2",
            "name": "GLM-5.2",
            "title": "GLM-5.2: Built for Long-Horizon Tasks",
            "lab": "zai",
            "date": "2026-06-16",
            "printed": "93.6",
            "where": "Full Benchmark Table, row 'GPQA-Diamond', column 'GPT-5.5'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/4fc1be3b62dac325.png"
      },
      {
        "id": "gpt-5-5:hle:no-tools:1",
        "model": "gpt-5-5",
        "model_name": "GPT-5.5",
        "model_date": "2026-04-23",
        "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 41.4,
        "unit": "%",
        "printed": "41.4%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.5",
        "source": "https://openai.com/index/introducing-gpt-5-5",
        "source_title": "Introducing GPT-5.5",
        "source_date": "2026-04-23",
        "where": "academic eval table, column 'GPT‑5.5', row 'Humanity's Last Exam (no tools)'",
        "own": true,
        "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://www.anthropic.com/claude-fable-5-mythos-5-system-card",
            "name": "Claude Fable 5 and Mythos 5 card",
            "title": "Claude Fable 5 and Claude Mythos 5 System Card",
            "lab": "anthropic",
            "date": "2026-06-09",
            "printed": "41.4",
            "where": "Table 8.1.A (p.251), row Humanity's Last Exam / No tools, column GPT-5.5"
          },
          {
            "url": "https://www.anthropic.com/claude-opus-4-8-system-card",
            "name": "Claude Opus 4.8 card",
            "title": "Claude Opus 4.8 System Card",
            "lab": "anthropic",
            "date": "2026-05-28",
            "printed": "41.4",
            "where": "Table 8.1.A (p.194), row Humanity's Last Exam / No tools, column GPT-5.5"
          },
          {
            "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": "41.4",
            "where": "Table 8.1.A (p.115), row Humanity's Last Exam / No tools, column GPT-5.5"
          },
          {
            "url": "https://www.anthropic.com/news/claude-opus-4-8",
            "name": "Claude Opus 4.8",
            "title": "Introducing Claude Opus 4.8",
            "lab": "anthropic",
            "date": "2026-05-28",
            "printed": "41.4%",
            "where": "benchmark table image, row Humanity's Last Exam (no tools), column GPT-5.5"
          },
          {
            "url": "https://www.kimi.ai/blog/kimi-k3",
            "name": "Kimi K3",
            "title": "Kimi K3: Open Frontier Intelligence",
            "lab": "moonshot",
            "date": "2026-07-16",
            "printed": "41.4",
            "where": "Full Benchmark Table, row HLE-Full ; column GPT 5.5 (xhigh)"
          },
          {
            "url": "https://z.ai/blog/glm-5.2",
            "name": "GLM-5.2",
            "title": "GLM-5.2: Built for Long-Horizon Tasks",
            "lab": "zai",
            "date": "2026-06-16",
            "printed": "41.4",
            "where": "Chart 'LLM Performance Evaluation' (all models at max thinking effort), panel 'Humanity's Last Exam', bar 'GPT-5.5' (lower label, not marked w/ Tools)"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/4fc1be3b62dac325.png"
      },
      {
        "id": "gpt-5-5:hle:with-tools:1",
        "model": "gpt-5-5",
        "model_name": "GPT-5.5",
        "model_date": "2026-04-23",
        "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "with-tools",
        "setting_name": "With tools",
        "score": 52.2,
        "unit": "%",
        "printed": "52.2%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.5",
        "source": "https://openai.com/index/introducing-gpt-5-5",
        "source_title": "Introducing GPT-5.5",
        "source_date": "2026-04-23",
        "where": "academic eval table, column 'GPT‑5.5', row 'Humanity's Last Exam (with tools)'",
        "own": true,
        "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://www.anthropic.com/claude-fable-5-mythos-5-system-card",
            "name": "Claude Fable 5 and Mythos 5 card",
            "title": "Claude Fable 5 and Claude Mythos 5 System Card",
            "lab": "anthropic",
            "date": "2026-06-09",
            "printed": "52.2",
            "where": "Table 8.1.A (p.251), row Humanity's Last Exam / With tools, column GPT-5.5"
          },
          {
            "url": "https://www.anthropic.com/claude-opus-4-8-system-card",
            "name": "Claude Opus 4.8 card",
            "title": "Claude Opus 4.8 System Card",
            "lab": "anthropic",
            "date": "2026-05-28",
            "printed": "52.2",
            "where": "Table 8.1.A (p.194), row Humanity's Last Exam / With tools, column GPT-5.5"
          },
          {
            "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": "52.2",
            "where": "Table 8.1.A (p.115), row Humanity's Last Exam / With tools, column GPT-5.5"
          },
          {
            "url": "https://www.anthropic.com/news/claude-opus-4-8",
            "name": "Claude Opus 4.8",
            "title": "Introducing Claude Opus 4.8",
            "lab": "anthropic",
            "date": "2026-05-28",
            "printed": "52.2%",
            "where": "benchmark table image, row Humanity's Last Exam (with tools), column GPT-5.5"
          },
          {
            "url": "https://www.kimi.ai/blog/kimi-k3",
            "name": "Kimi K3",
            "title": "Kimi K3: Open Frontier Intelligence",
            "lab": "moonshot",
            "date": "2026-07-16",
            "printed": "52.2",
            "where": "Full Benchmark Table, row HLE-Full w/ tools ; column GPT 5.5 (xhigh)"
          },
          {
            "url": "https://z.ai/blog/glm-5.2",
            "name": "GLM-5.2",
            "title": "GLM-5.2: Built for Long-Horizon Tasks",
            "lab": "zai",
            "date": "2026-06-16",
            "printed": "52.2",
            "where": "Full Benchmark Table, row 'HLE w/ Tools (* = full set for Opus 4.8, GPT-5.5, Gemini 3.1 Pro)', column 'GPT-5.5'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/4fc1be3b62dac325.png"
      },
      {
        "id": "gpt-5-5:arc-agi-2:default:1",
        "model": "gpt-5-5",
        "model_name": "GPT-5.5",
        "model_date": "2026-04-23",
        "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
        "benchmark": "arc-agi-2",
        "benchmark_name": "ARC-AGI-2",
        "setting": "default",
        "setting_name": "",
        "score": 85.0,
        "unit": "%",
        "printed": "85.0%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.5",
        "source": "https://openai.com/index/introducing-gpt-5-5",
        "source_title": "Introducing GPT-5.5",
        "source_date": "2026-04-23",
        "where": "abstract reasoning eval table, column 'GPT‑5.5', row 'ARC-AGI-2 (Verified)'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "ARC Prize Verified",
        "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/44368b4c5f191a53.png"
      },
      {
        "id": "gpt-5-5:browsecomp:default:1",
        "model": "gpt-5-5",
        "model_name": "GPT-5.5",
        "model_date": "2026-04-23",
        "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
        "benchmark": "browsecomp",
        "benchmark_name": "BrowseComp",
        "setting": "default",
        "setting_name": "",
        "score": 84.4,
        "unit": "%",
        "printed": "84.4%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.5",
        "source": "https://openai.com/index/introducing-gpt-5-5",
        "source_title": "Introducing GPT-5.5",
        "source_date": "2026-04-23",
        "where": "tool-use eval table (columns GPT‑5.5 / GPT‑5.4 / GPT‑5.5 Pro / GPT‑5.4 Pro / Claude Opus 4.7 / Gemini 3.1 Pro), row 'BrowseComp', column 'GPT‑5.5'",
        "own": true,
        "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://www.anthropic.com/claude-fable-5-mythos-5-system-card",
            "name": "Claude Fable 5 and Mythos 5 card",
            "title": "Claude Fable 5 and Claude Mythos 5 System Card",
            "lab": "anthropic",
            "date": "2026-06-09",
            "printed": "84.4",
            "where": "Table 8.1.A (p.251), row BrowseComp, column GPT-5.5"
          },
          {
            "url": "https://www.anthropic.com/claude-opus-4-8-system-card",
            "name": "Claude Opus 4.8 card",
            "title": "Claude Opus 4.8 System Card",
            "lab": "anthropic",
            "date": "2026-05-28",
            "printed": "84.4",
            "where": "Table 8.1.A (p.194), row BrowseComp, column GPT-5.5"
          },
          {
            "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": "84.4",
            "where": "Table 8.1.A (p.115), row BrowseComp, column GPT-5.5"
          },
          {
            "url": "https://www.minimax.io/blog/minimax-m3",
            "name": "MiniMax M3",
            "title": "MiniMax M3: Frontier Coding, 1M Context, Native Multimodality — All in One Model",
            "lab": "minimax",
            "date": "2026-06-01",
            "printed": "84.4",
            "where": "Bar chart image (10 panels); panel 'BrowseComp'; value label on GPT 5.5 bar"
          },
          {
            "url": "https://www.kimi.ai/blog/kimi-k3",
            "name": "Kimi K3",
            "title": "Kimi K3: Open Frontier Intelligence",
            "lab": "moonshot",
            "date": "2026-07-16",
            "printed": "84.4",
            "where": "Full Benchmark Table, row BrowseComp ; column GPT 5.5 (xhigh)"
          },
          {
            "url": "https://openai.com/index/gpt-5-6",
            "name": "GPT-5.6",
            "title": "GPT-5.6: Frontier intelligence that scales with your ambition",
            "lab": "openai",
            "date": "2026-07-09",
            "printed": "84.4%",
            "where": "table 3 \"BrowseComp\" row, \"GPT-5.5\" column"
          },
          {
            "url": "https://openai.com/index/builders-guide-to-gpt-5-6",
            "name": "OpenAI post",
            "title": "The builder’s guide to GPT‑5.6",
            "lab": "openai",
            "date": "2026-08-13",
            "printed": "84.36%",
            "where": "sentence: \"GPT-5.5 (Extra High) scored 84.36% on this benchmark [BrowseComp]\""
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/7f009782c6b95376.png"
      },
      {
        "id": "gpt-5-5:swe-bench-pro:default:1",
        "model": "gpt-5-5",
        "model_name": "GPT-5.5",
        "model_date": "2026-04-23",
        "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
        "benchmark": "swe-bench-pro",
        "benchmark_name": "SWE-bench Pro",
        "setting": "default",
        "setting_name": "",
        "score": 58.6,
        "unit": "%",
        "printed": "58.6%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.5",
        "source": "https://openai.com/index/introducing-gpt-5-5",
        "source_title": "Introducing GPT-5.5",
        "source_date": "2026-04-23",
        "where": "coding eval table, row 'SWE-Bench Pro (Public) *', column 'GPT‑5.5'",
        "own": true,
        "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://www.anthropic.com/claude-fable-5-mythos-5-system-card",
            "name": "Claude Fable 5 and Mythos 5 card",
            "title": "Claude Fable 5 and Claude Mythos 5 System Card",
            "lab": "anthropic",
            "date": "2026-06-09",
            "printed": "58.6",
            "where": "Table 8.1.A (p.251), row SWE-bench Pro, column GPT-5.5"
          },
          {
            "url": "https://www.anthropic.com/claude-opus-4-8-system-card",
            "name": "Claude Opus 4.8 card",
            "title": "Claude Opus 4.8 System Card",
            "lab": "anthropic",
            "date": "2026-05-28",
            "printed": "58.6",
            "where": "Table 8.1.A (p.194), row SWE-bench Pro, column GPT-5.5"
          },
          {
            "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": "58.6",
            "where": "Table 8.1.A (p.115), row SWE-bench Pro, column GPT-5.5"
          },
          {
            "url": "https://www.anthropic.com/news/claude-opus-4-8",
            "name": "Claude Opus 4.8",
            "title": "Introducing Claude Opus 4.8",
            "lab": "anthropic",
            "date": "2026-05-28",
            "printed": "58.6%",
            "where": "benchmark table image, row SWE-Bench Pro, column GPT-5.5"
          },
          {
            "url": "https://www.minimax.io/blog/minimax-m3",
            "name": "MiniMax M3",
            "title": "MiniMax M3: Frontier Coding, 1M Context, Native Multimodality — All in One Model",
            "lab": "minimax",
            "date": "2026-06-01",
            "printed": "58.6",
            "where": "Bar chart image (10 panels); panel 'SWE Bench Pro'; value label on GPT 5.5 bar"
          },
          {
            "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-5/",
            "name": "Gemini 3.5 Flash",
            "title": "Gemini 3.5: frontier intelligence with action",
            "lab": "google",
            "date": "2026-05-19",
            "printed": "58.6%",
            "where": "Benchmark table image; row 'SWE-Bench Pro (Public), Single attempt'; column gpt-5-5"
          },
          {
            "url": "https://z.ai/blog/glm-5.2",
            "name": "GLM-5.2",
            "title": "GLM-5.2: Built for Long-Horizon Tasks",
            "lab": "zai",
            "date": "2026-06-16",
            "printed": "58.6",
            "where": "Full Benchmark Table, row 'SWE-bench Pro', column 'GPT-5.5'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/85f894b0a6f836f5.png"
      },
      {
        "id": "gpt-5-5:terminal-bench-2-0:default:1",
        "model": "gpt-5-5",
        "model_name": "GPT-5.5",
        "model_date": "2026-04-23",
        "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
        "benchmark": "terminal-bench-2-0",
        "benchmark_name": "Terminal-Bench 2.0",
        "setting": "default",
        "setting_name": "",
        "score": 82.7,
        "unit": "%",
        "printed": "82.7%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.5",
        "source": "https://openai.com/index/introducing-gpt-5-5",
        "source_title": "Introducing GPT-5.5",
        "source_date": "2026-04-23",
        "where": "headline eval table (HTML, first table), row 'Terminal-Bench 2.0', column 'GPT‑5.5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "",
        "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/85f894b0a6f836f5.png"
      },
      {
        "id": "gpt-5-5:osworld-verified:default:1",
        "model": "gpt-5-5",
        "model_name": "GPT-5.5",
        "model_date": "2026-04-23",
        "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
        "benchmark": "osworld-verified",
        "benchmark_name": "OSWorld-Verified",
        "setting": "default",
        "setting_name": "",
        "score": 78.7,
        "unit": "%",
        "printed": "78.7%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.5",
        "source": "https://openai.com/index/introducing-gpt-5-5",
        "source_title": "Introducing GPT-5.5",
        "source_date": "2026-04-23",
        "where": "headline eval table (HTML, first table), row 'OSWorld-Verified', column 'GPT‑5.5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-18",
        "check": "human",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://www.anthropic.com/claude-fable-5-mythos-5-system-card",
            "name": "Claude Fable 5 and Mythos 5 card",
            "title": "Claude Fable 5 and Claude Mythos 5 System Card",
            "lab": "anthropic",
            "date": "2026-06-09",
            "printed": "78.7",
            "where": "Table 8.1.A (p.251), row OSWorld-Verified, column GPT-5.5"
          },
          {
            "url": "https://www.anthropic.com/claude-opus-4-8-system-card",
            "name": "Claude Opus 4.8 card",
            "title": "Claude Opus 4.8 System Card",
            "lab": "anthropic",
            "date": "2026-05-28",
            "printed": "78.7",
            "where": "Table 8.1.A (p.194), row OSWorld-Verified, column GPT-5.5"
          },
          {
            "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": "78.7",
            "where": "Table 8.1.A (p.115), row OSWorld-Verified, column GPT-5.5"
          },
          {
            "url": "https://www.anthropic.com/news/claude-opus-4-8",
            "name": "Claude Opus 4.8",
            "title": "Introducing Claude Opus 4.8",
            "lab": "anthropic",
            "date": "2026-05-28",
            "printed": "78.7%",
            "where": "benchmark table image, row OSWorld-Verified, column GPT-5.5"
          },
          {
            "url": "https://www.minimax.io/blog/minimax-m3",
            "name": "MiniMax M3",
            "title": "MiniMax M3: Frontier Coding, 1M Context, Native Multimodality — All in One Model",
            "lab": "minimax",
            "date": "2026-06-01",
            "printed": "78.7",
            "where": "Bar chart image (10 panels); panel 'OSWorld-verified'; value label on GPT 5.5 bar"
          },
          {
            "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-5/",
            "name": "Gemini 3.5 Flash",
            "title": "Gemini 3.5: frontier intelligence with action",
            "lab": "google",
            "date": "2026-05-19",
            "printed": "78.7%",
            "where": "Benchmark table image; row 'OSWorld-Verified'; column gpt-5-5"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/b1a005f7e1202b4c.png"
      },
      {
        "id": "gpt-5-5:arc-agi-1:default:1",
        "model": "gpt-5-5",
        "model_name": "GPT-5.5",
        "model_date": "2026-04-23",
        "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
        "benchmark": "arc-agi-1",
        "benchmark_name": "ARC-AGI-1",
        "setting": "default",
        "setting_name": "",
        "score": 95.0,
        "unit": "%",
        "printed": "95.0%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.5",
        "source": "https://openai.com/index/introducing-gpt-5-5",
        "source_title": "Introducing GPT-5.5",
        "source_date": "2026-04-23",
        "where": "detail eval table (HTML, last table), row 'ARC-AGI-1 (Verified)', column 'GPT‑5.5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "ARC Prize Verified",
        "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/44368b4c5f191a53.png"
      },
      {
        "id": "gpt-5-5:automationbench:default:1",
        "model": "gpt-5-5",
        "model_name": "GPT-5.5",
        "model_date": "2026-04-23",
        "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
        "benchmark": "automationbench",
        "benchmark_name": "AutomationBench",
        "setting": "default",
        "setting_name": "",
        "score": 12.9,
        "unit": "%",
        "printed": "12.9%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT-5.5",
        "source": "https://openai.com/index/gpt-5-6",
        "source_title": "GPT-5.6: Frontier intelligence that scales with your ambition",
        "source_date": "2026-07-09",
        "where": "tool-use table (columns GPT-5.6 Sol, Terra, Luna, GPT-5.5, ...), row 'AutomationBench', column 'GPT-5.5'",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "",
        "caveat": "",
        "human_verified": false,
        "human_verified_on": null,
        "check": "corroborated",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://www.anthropic.com/news/claude-fable-5-mythos-5",
            "name": "Claude Fable 5 and Mythos 5",
            "title": "Claude Fable 5 and Claude Mythos 5",
            "lab": "anthropic",
            "date": "2026-06-09",
            "printed": "12.9%",
            "where": "Benchmark table image, row \"Tool use AutomationBench\", column \"GPT 5.5\""
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/90318aa23c794381.png"
      },
      {
        "id": "gpt-5-5:arc-agi-3:default:1",
        "model": "gpt-5-5",
        "model_name": "GPT-5.5",
        "model_date": "2026-04-23",
        "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
        "benchmark": "arc-agi-3",
        "benchmark_name": "ARC-AGI-3",
        "setting": "default",
        "setting_name": "",
        "score": 0.43,
        "unit": "%",
        "printed": "0.43%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT-5.5",
        "source": "https://openai.com/index/gpt-5-6",
        "source_title": "GPT-5.6: Frontier intelligence that scales with your ambition",
        "source_date": "2026-07-09",
        "where": "abstract-reasoning table (last table), row 'ARC-AGI-3⁷', column 'GPT-5.5'",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "",
        "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/669da4e40aae1378.png"
      },
      {
        "id": "gpt-5-5:swe-bench-pro:default:2",
        "model": "gpt-5-5",
        "model_name": "GPT-5.5",
        "model_date": "2026-04-23",
        "event": "https://frontierlog.ai/event/2026-04-23-gpt-5-5",
        "benchmark": "swe-bench-pro",
        "benchmark_name": "SWE-bench Pro",
        "setting": "default",
        "setting_name": "",
        "score": 59.4,
        "unit": "%",
        "printed": "59.4%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT-5.5",
        "source": "https://openai.com/index/gpt-5-6",
        "source_title": "GPT-5.6: Frontier intelligence that scales with your ambition",
        "source_date": "2026-07-09",
        "where": "coding table (columns GPT-5.6 Sol, Sol Ultra, Terra, Luna, GPT-5.5, ...), row 'SWE-Bench Pro', column 'GPT-5.5'",
        "own": false,
        "primary": false,
        "parallel": false,
        "notes": "",
        "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/9e2ca9dd2373f9f9.png"
      }
    ]
  }
}