{
  "api_version": "1",
  "updated": "2026-10-26",
  "latest_event": "2026-10-24",
  "rights": "© Frontier Log, all rights reserved.",
  "model": {
    "slug": "gpt-5-1",
    "name": "GPT-5.1",
    "lab": "openai",
    "date": "2025-11-12",
    "source": "https://openai.com/index/gpt-5-1",
    "source_title": "GPT-5.1: A smarter, more conversational ChatGPT",
    "flagship": true,
    "event": "https://frontierlog.ai/event/2025-11-12-gpt-5-1",
    "releases": [
      "https://frontierlog.ai/event/2025-11-13-gpt-5-1-api"
    ],
    "variant_of": null,
    "made_by": "OpenAI",
    "url": "https://frontierlog.ai/model/gpt-5-1",
    "results": [
      {
        "id": "gpt-5-1:swe-bench-verified:default:1",
        "model": "gpt-5-1",
        "model_name": "GPT-5.1",
        "model_date": "2025-11-12",
        "event": "https://frontierlog.ai/event/2025-11-12-gpt-5-1",
        "benchmark": "swe-bench-verified",
        "benchmark_name": "SWE-bench Verified",
        "setting": "default",
        "setting_name": "",
        "score": 76.3,
        "unit": "%",
        "printed": "76.3%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.1 Thinking",
        "source": "https://openai.com/index/introducing-gpt-5-2",
        "source_title": "Introducing GPT-5.2",
        "source_date": "2025-12-11",
        "where": "first comparison table, column 'GPT‑5.1 Thinking', row 'SWE-bench Verified'",
        "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://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": "76.3%",
            "where": "Table 2.3.A (p.19), row SWE-bench Verified, column GPT-5.1"
          },
          {
            "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": "76.3%",
            "where": "Benchmark table image, row 'Agentic coding SWE-bench Verified', column 'GPT-5.1', upper value (lower value is GPT-5.1-Codex-Max)"
          },
          {
            "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": "76.3%",
            "where": "'SWE-bench Verified Score Evolution' line chart image; point label 'GPT-5.1 76.3%'"
          },
          {
            "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": "76.3%",
            "where": "Benchmark table image; row 'SWE-Bench Verified, Single attempt'; column gpt-5-1"
          },
          {
            "url": "https://openai.com/index/gpt-5-1-for-developers",
            "name": "GPT-5.1 in the API",
            "title": "Introducing GPT-5.1 for developers",
            "lab": "openai",
            "date": "2025-11-13",
            "printed": "76.3%",
            "where": "table 0 \"SWE-bench Verified(all 500 problems)\" row, \"GPT-5.1 (high)\" column"
          },
          {
            "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": "76.3",
            "where": "Benchmark table, row 'SWE-bench Verified', column 'GPT-5.1 High'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/da9899c90caed7cc.png"
      },
      {
        "id": "gpt-5-1:gpqa-diamond:default:1",
        "model": "gpt-5-1",
        "model_name": "GPT-5.1",
        "model_date": "2025-11-12",
        "event": "https://frontierlog.ai/event/2025-11-12-gpt-5-1",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 88.1,
        "unit": "%",
        "printed": "88.1%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.1 Thinking",
        "source": "https://openai.com/index/introducing-gpt-5-2",
        "source_title": "Introducing GPT-5.2",
        "source_date": "2025-12-11",
        "where": "first comparison table, column 'GPT‑5.1 Thinking', row 'GPQA Diamond (no tools)'",
        "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://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": "88.1%",
            "where": "Table 2.3.A (p.19), row GPQA Diamond, column GPT-5.1"
          },
          {
            "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": "88.1%",
            "where": "Benchmark table image, row 'Graduate-level reasoning GPQA Diamond', column 'GPT-5.1'"
          },
          {
            "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": "88.1%",
            "where": "Benchmark table image; row 'GPQA Diamond, No tools'; column gpt-5-1"
          },
          {
            "url": "https://openai.com/index/gpt-5-1-for-developers",
            "name": "GPT-5.1 in the API",
            "title": "Introducing GPT-5.1 for developers",
            "lab": "openai",
            "date": "2025-11-13",
            "printed": "88.1%",
            "where": "table 0 \"GPQA Diamond (no tools)\" row, \"GPT-5.1 (high)\" column"
          },
          {
            "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": "88.1",
            "where": "Benchmark table, row 'GPQA-Diamond', column 'GPT-5.1 High'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/22817d0b2b0a2432.png"
      },
      {
        "id": "gpt-5-1:arc-agi-2:default:1",
        "model": "gpt-5-1",
        "model_name": "GPT-5.1",
        "model_date": "2025-11-12",
        "event": "https://frontierlog.ai/event/2025-11-12-gpt-5-1",
        "benchmark": "arc-agi-2",
        "benchmark_name": "ARC-AGI-2",
        "setting": "default",
        "setting_name": "",
        "score": 17.6,
        "unit": "%",
        "printed": "17.6%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.1 Thinking",
        "source": "https://openai.com/index/introducing-gpt-5-2",
        "source_title": "Introducing GPT-5.2",
        "source_date": "2025-12-11",
        "where": "first comparison table, column 'GPT‑5.1 Thinking', row 'ARC-AGI-2 (Verified)'",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "ARC Prize Verified",
        "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-4-5-system-card",
            "name": "Claude Opus 4.5 card",
            "title": "Claude Opus 4.5 System Card",
            "lab": "anthropic",
            "date": "2025-11-24",
            "printed": "17.6%",
            "where": "Table 2.3.A (p.19), row ARC-AGI-2 (Verified), column GPT-5.1"
          },
          {
            "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": "17.6%",
            "where": "Benchmark table image, row 'Novel problem solving ARC-AGI-2 (Verified)', column 'GPT-5.1'"
          },
          {
            "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": "17.6%",
            "where": "Benchmark table image; row 'ARC-AGI-2, ARC Prize Verified'; column gpt-5-1"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/5c5b6d32d6479128.png"
      },
      {
        "id": "gpt-5-1:hle:no-tools:1",
        "model": "gpt-5-1",
        "model_name": "GPT-5.1",
        "model_date": "2025-11-12",
        "event": "https://frontierlog.ai/event/2025-11-12-gpt-5-1",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 25.7,
        "unit": "%",
        "printed": "25.7%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT-5.1 Thinking",
        "source": "https://openai.com/index/introducing-gpt-5-2",
        "source_title": "Introducing GPT-5.2",
        "source_date": "2025-12-11",
        "where": "science and math eval table, column 'GPT-5.1 Thinking', row 'HLE (no tools)'",
        "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://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": "25.7",
            "where": "Benchmark table, row 'HLE (separate from 'HLE w/ Tools' row)', column 'GPT-5.1 High'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/22817d0b2b0a2432.png"
      },
      {
        "id": "gpt-5-1:hle:with-tools:1",
        "model": "gpt-5-1",
        "model_name": "GPT-5.1",
        "model_date": "2025-11-12",
        "event": "https://frontierlog.ai/event/2025-11-12-gpt-5-1",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "with-tools",
        "setting_name": "With tools",
        "score": 42.7,
        "unit": "%",
        "printed": "42.7%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT-5.1 Thinking",
        "source": "https://openai.com/index/introducing-gpt-5-2",
        "source_title": "Introducing GPT-5.2",
        "source_date": "2025-12-11",
        "where": "science and math eval table, column 'GPT-5.1 Thinking', row 'HLE (w/ search, Python)'",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "search and Python",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-17",
        "check": "human",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "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": "42.7",
            "where": "Benchmark table, row 'HLE w/ Tools', column 'GPT-5.1 High'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/22817d0b2b0a2432.png"
      },
      {
        "id": "gpt-5-1:swe-bench-pro:default:1",
        "model": "gpt-5-1",
        "model_name": "GPT-5.1",
        "model_date": "2025-11-12",
        "event": "https://frontierlog.ai/event/2025-11-12-gpt-5-1",
        "benchmark": "swe-bench-pro",
        "benchmark_name": "SWE-bench Pro",
        "setting": "default",
        "setting_name": "",
        "score": 50.8,
        "unit": "%",
        "printed": "50.8%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT-5.1 Thinking",
        "source": "https://openai.com/index/introducing-gpt-5-2",
        "source_title": "Introducing GPT-5.2",
        "source_date": "2025-12-11",
        "where": "coding table (columns GPT-5.2 Thinking, GPT-5.2 Pro, GPT-5.1 Thinking), row 'SWE-Bench Pro, Public', column 'GPT-5.1 Thinking'",
        "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://openai.com/index/introducing-gpt-5-2-codex",
            "name": "GPT-5.2-Codex",
            "title": "Introducing GPT-5.2-Codex",
            "lab": "openai",
            "date": "2025-12-18",
            "printed": "50.8%",
            "where": "SWE-Bench Pro chart, bar 'GPT-5.1' (bars: GPT-5.2-Codex 56.4%, GPT-5.2 55.6%, GPT-5.1 50.8%); effort not stated"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/da9899c90caed7cc.png"
      },
      {
        "id": "gpt-5-1:browsecomp:default:1",
        "model": "gpt-5-1",
        "model_name": "GPT-5.1",
        "model_date": "2025-11-12",
        "event": "https://frontierlog.ai/event/2025-11-12-gpt-5-1",
        "benchmark": "browsecomp",
        "benchmark_name": "BrowseComp",
        "setting": "default",
        "setting_name": "",
        "score": 50.8,
        "unit": "%",
        "printed": "50.8%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT-5.1 Thinking",
        "source": "https://openai.com/index/introducing-gpt-5-2",
        "source_title": "Introducing GPT-5.2",
        "source_date": "2025-12-11",
        "where": "agentic tool-use table (columns GPT-5.2 Thinking, GPT-5.2 Pro, GPT-5.1 Thinking), row 'BrowseComp', column 'GPT-5.1 Thinking'",
        "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://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": "50.8",
            "where": "Benchmark table, row 'BrowseComp', column 'GPT-5.1 High'. No provenance footnote."
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/a408379f5cd396d1.png"
      }
    ]
  }
}