{
  "api_version": "1",
  "updated": "2026-10-26",
  "latest_event": "2026-10-24",
  "rights": "© Frontier Log, all rights reserved.",
  "model": {
    "slug": "gpt-5-6-sol",
    "name": "GPT-5.6 Sol",
    "lab": "openai",
    "date": "2026-07-09",
    "source": "https://openai.com/index/gpt-5-6",
    "source_title": "GPT-5.6: Frontier intelligence that scales with your ambition",
    "flagship": true,
    "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
    "releases": [
      "https://frontierlog.ai/event/2026-06-26-gpt-5-6-sol-preview",
      "https://frontierlog.ai/event/2026-08-06-gpt-5-6-luna-free-users"
    ],
    "variant_of": null,
    "made_by": "OpenAI",
    "url": "https://frontierlog.ai/model/gpt-5-6-sol",
    "results": [
      {
        "id": "gpt-5-6-sol:gpqa-diamond:default:1",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 94.6,
        "unit": "%",
        "printed": "94.6%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.6 Sol",
        "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": "academic eval table, column 'GPT‑5.6 Sol', 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-6-astra",
            "name": "GPT-6 Astra",
            "title": "GPT-6 Astra: A new generation of intelligence",
            "lab": "openai",
            "date": "2026-09-03",
            "printed": "94.6%",
            "where": "table 3 \"GPQA Diamond\" row, \"GPT-5.6 Sol\" column"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/99df6dac1337da2b.png"
      },
      {
        "id": "gpt-5-6-sol:arc-agi-2:default:1",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "arc-agi-2",
        "benchmark_name": "ARC-AGI-2",
        "setting": "default",
        "setting_name": "",
        "score": 92.5,
        "unit": "%",
        "printed": "92.5%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.6 Sol",
        "source": "https://openai.com/index/gpt-6-astra",
        "source_title": "GPT-6 Astra: A new generation of intelligence",
        "source_date": "2026-09-03",
        "where": "'Abstract reasoning' table, column 'GPT‑5.6 Sol', row 'ARC-AGI-2'",
        "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-fable-5-1-mythos-5-1-system-card",
            "name": "Claude Fable 5.1 and Mythos 5.1 card",
            "title": "Claude Fable 5.1 and Claude Mythos 5.1 System Card",
            "lab": "anthropic",
            "date": "2026-09-01",
            "printed": "92.5%",
            "where": "Table 8.1.A (p.167), row ARC-AGI-2, column GPT-5.6 Sol"
          },
          {
            "url": "https://www.anthropic.com/claude-opus-5-system-card",
            "name": "Claude Opus 5 card",
            "title": "Claude Opus 5 System Card",
            "lab": "anthropic",
            "date": "2026-07-24",
            "printed": "92.5",
            "where": "Table 8.1.A (p.153), row ARC-AGI-2, column GPT 5.6 Sol"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/6917e15279360166.png"
      },
      {
        "id": "gpt-5-6-sol:browsecomp:default:1",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "browsecomp",
        "benchmark_name": "BrowseComp",
        "setting": "default",
        "setting_name": "",
        "score": 90.4,
        "unit": "%",
        "printed": "90.4%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.6 Sol",
        "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": "eval table with columns GPT‑5.6 Sol / Sol Ultra / Terra / Luna / …, row 'BrowseComp', column 'GPT‑5.6 Sol'",
        "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-opus-5-system-card",
            "name": "Claude Opus 5 card",
            "title": "Claude Opus 5 System Card",
            "lab": "anthropic",
            "date": "2026-07-24",
            "printed": "90.4",
            "where": "Table 8.1.A (p.152), row BrowseComp, column GPT 5.6 Sol"
          },
          {
            "url": "https://www.anthropic.com/news/claude-opus-5",
            "name": "Claude Opus 5",
            "title": "Introducing Claude Opus 5",
            "lab": "anthropic",
            "date": "2026-07-24",
            "printed": "90.4%",
            "where": "benchmark table image, row BrowseComp, column GPT-5.6 Sol"
          },
          {
            "url": "https://www.kimi.ai/blog/kimi-k3",
            "name": "Kimi K3",
            "title": "Kimi K3: Open Frontier Intelligence",
            "lab": "moonshot",
            "date": "2026-07-16",
            "printed": "90.4",
            "where": "Full Benchmark Table, row BrowseComp ; column GPT 5.6 Sol (max)"
          },
          {
            "url": "https://openai.com/index/gpt-6-astra",
            "name": "GPT-6 Astra",
            "title": "GPT-6 Astra: A new generation of intelligence",
            "lab": "openai",
            "date": "2026-09-03",
            "printed": "90.4%",
            "where": "table 1 \"BrowseComp\" row, \"GPT-5.6 Sol\" column"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/2eabd1ed1ba3ff72.png"
      },
      {
        "id": "gpt-5-6-sol:swe-bench-pro:default:1",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "swe-bench-pro",
        "benchmark_name": "SWE-bench Pro",
        "setting": "default",
        "setting_name": "",
        "score": 64.6,
        "unit": "%",
        "printed": "64.6%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.6 Sol",
        "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 eval table (columns GPT‑5.6 Sol, Sol Ultra, Terra, Luna, GPT‑5.5, …), row 'SWE-Bench Pro', column 'GPT‑5.6 Sol'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "",
        "caveat": "The source doesn't say whether this is the public or the commercial set",
        "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-5-1-mythos-5-1-system-card",
            "name": "Claude Fable 5.1 and Mythos 5.1 card",
            "title": "Claude Fable 5.1 and Claude Mythos 5.1 System Card",
            "lab": "anthropic",
            "date": "2026-09-01",
            "printed": "64.6",
            "where": "Table 8.1.A (p.167), row SWE-bench Pro, column GPT-5.6 Sol"
          },
          {
            "url": "https://www.anthropic.com/claude-opus-5-system-card",
            "name": "Claude Opus 5 card",
            "title": "Claude Opus 5 System Card",
            "lab": "anthropic",
            "date": "2026-07-24",
            "printed": "64.6",
            "where": "Table 8.1.A (p.152), row SWE-bench Pro, column GPT 5.6 Sol"
          },
          {
            "url": "https://qwen.ai/blog?id=qwen3.8",
            "name": "Qwen3.8-Max",
            "title": "Qwen3.8-Max: A New Bar for Coding and Cowork",
            "lab": "alibaba",
            "date": "2026-08-03",
            "printed": "64.6",
            "where": "Performance chart image (performance.png), panel SWE-Pro, value label on the GPT5.6 Sol (max) bar"
          },
          {
            "url": "https://thinkingmachines.ai/news/introducing-inkling/",
            "name": "Inkling",
            "title": "Inkling: Our Open-Weights Model",
            "lab": "thinking-machines",
            "date": "2026-07-15",
            "printed": "64.6%",
            "where": "'Benchmarking Inkling' main table, row 'SWEBench Pro Public', column 'GPT 5.6 Sol (max/xhigh)'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/9e2ca9dd2373f9f9.png"
      },
      {
        "id": "gpt-5-6-sol:deepswe-1-1:default:1",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "deepswe-1-1",
        "benchmark_name": "DeepSWE v1.1",
        "setting": "default",
        "setting_name": "",
        "score": 72.7,
        "unit": "%",
        "printed": "72.7%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.6 Sol",
        "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 eval table (columns GPT‑5.6 Sol, Sol Ultra, Terra, Luna, GPT‑5.5, …), row 'DeepSWE v1.1', column 'GPT‑5.6 Sol'",
        "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-opus-5-system-card",
            "name": "Claude Opus 5 card",
            "title": "Claude Opus 5 System Card",
            "lab": "anthropic",
            "date": "2026-07-24",
            "printed": "72.7",
            "where": "Table 8.1.A (p.152), row DeepSWE v1.1, column GPT 5.6 Sol"
          },
          {
            "url": "https://www.anthropic.com/news/claude-opus-5",
            "name": "Claude Opus 5",
            "title": "Introducing Claude Opus 5",
            "lab": "anthropic",
            "date": "2026-07-24",
            "printed": "72.7%",
            "where": "benchmark table image, row DeepSWE v1.1, column GPT-5.6 Sol"
          },
          {
            "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/",
            "name": "Gemini 3.8 Flash and 3.8 Flash Cyber",
            "title": "Introducing Gemini 3.8 Flash and 3.8 Flash Cyber",
            "lab": "google",
            "date": "2026-09-02",
            "printed": "72.7%",
            "where": "Benchmark table image; row 'DeepSWE v1.1'; column gpt-5-6-sol"
          },
          {
            "url": "https://openai.com/index/gpt-6-astra",
            "name": "GPT-6 Astra",
            "title": "GPT-6 Astra: A new generation of intelligence",
            "lab": "openai",
            "date": "2026-09-03",
            "printed": "72.7%",
            "where": "table 2 \"DeepSWE v1.1\" row, \"GPT-5.6 Sol\" column"
          },
          {
            "url": "https://z.ai/blog/glm-5.3",
            "name": "GLM-5.3",
            "title": "GLM-5.3: Frontier Coding with Emergent Cyber Capabilities",
            "lab": "zai",
            "date": "2026-08-14",
            "printed": "72.7",
            "where": "'Performance across comparison models' table, row 'DeepSWE v1.1', column 'GPT-5.6 Sol'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/9e2ca9dd2373f9f9.png"
      },
      {
        "id": "gpt-5-6-sol:terminal-bench-4-0:default:1",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "terminal-bench-4-0",
        "benchmark_name": "Terminal-Bench 4.0",
        "setting": "default",
        "setting_name": "",
        "score": 37.3,
        "unit": "%",
        "printed": "37.3%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.6 Sol",
        "source": "https://openai.com/index/gpt-6-astra",
        "source_title": "GPT-6 Astra: A new generation of intelligence",
        "source_date": "2026-09-03",
        "where": "Coding table, row 'Terminal-Bench 4.0', column 'GPT‑5.6 Sol'",
        "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-fable-and-mythos-5-1",
            "name": "Claude Fable 5.1 and Mythos 5.1",
            "title": "Introducing Claude Fable 5.1 and Claude Mythos 5.1",
            "lab": "anthropic",
            "date": "2026-09-01",
            "printed": "37.3%",
            "where": "HTML benchmark table 'Here, you can see how Fable 5.1 compares', row Terminal-Bench 4.0, column GPT-5.6 Sol"
          },
          {
            "url": "https://www.anthropic.com/claude-opus-5-5",
            "name": "Claude Opus 5.5",
            "title": "Introducing Claude Opus 5.5",
            "lab": "anthropic",
            "date": "2026-09-22",
            "printed": "37.3%",
            "where": "HTML benchmark table under 'Performance and cost-effectiveness', row Terminal-Bench 4.0¹, column GPT-5.6 Sol (as reported by OpenAI)"
          },
          {
            "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/",
            "name": "Gemini 3.8 Flash and 3.8 Flash Cyber",
            "title": "Introducing Gemini 3.8 Flash and 3.8 Flash Cyber",
            "lab": "google",
            "date": "2026-09-02",
            "printed": "37.3%",
            "where": "Benchmark table image; row 'Terminal-bench 4.0'; column gpt-5-6-sol"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/4de8f321fbadf8b8.png"
      },
      {
        "id": "gpt-5-6-sol:terminal-bench-2-1:default:1",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "terminal-bench-2-1",
        "benchmark_name": "Terminal-Bench 2.1",
        "setting": "default",
        "setting_name": "",
        "score": 88.8,
        "unit": "%",
        "printed": "88.8%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.6 Sol",
        "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 eval table (HTML), row 'Terminal-Bench 2.1', column 'GPT‑5.6 Sol'",
        "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://api-docs.deepseek.com/news/news260910/",
            "name": "DeepSeek-V4.1-Flash",
            "title": "DeepSeek-V4.1-Flash: Smarter, Faster, More Efficient",
            "lab": "deepseek",
            "date": "2026-09-10",
            "printed": "88.8",
            "where": "Benchmark table image; row 'Terminal-Bench 2.1'; column 'GPT 5.6-Sol'"
          },
          {
            "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/",
            "name": "Gemini 3.8 Flash and 3.8 Flash Cyber",
            "title": "Introducing Gemini 3.8 Flash and 3.8 Flash Cyber",
            "lab": "google",
            "date": "2026-09-02",
            "printed": "88.8%",
            "where": "Benchmark table image; row 'Terminal-bench 2.1'; column gpt-5-6-sol"
          },
          {
            "url": "https://www.kimi.ai/blog/kimi-k3",
            "name": "Kimi K3",
            "title": "Kimi K3: Open Frontier Intelligence",
            "lab": "moonshot",
            "date": "2026-07-16",
            "printed": "88.8",
            "where": "Full Benchmark Table, row Terminal Bench 2.1 ; column GPT 5.6 Sol (max)"
          },
          {
            "url": "https://openai.com/index/previewing-gpt-5-6-sol",
            "name": "GPT-5.6 Sol preview",
            "title": "Previewing GPT-5.6 Sol: a next-generation model",
            "lab": "openai",
            "date": "2026-06-26",
            "printed": "88.8%",
            "where": "chart \"TerminalBench 2.1\" (Coding pass@1), value label on \"GPT-5.6 Sol\" bar"
          },
          {
            "url": "https://qwen.ai/blog?id=qwen3.8",
            "name": "Qwen3.8-Max",
            "title": "Qwen3.8-Max: A New Bar for Coding and Cowork",
            "lab": "alibaba",
            "date": "2026-08-03",
            "printed": "88.8",
            "where": "Performance chart image (performance.png), panel TerminalBench-2.1, value label on the GPT5.6 Sol (max) bar"
          },
          {
            "url": "https://z.ai/blog/glm-5.3",
            "name": "GLM-5.3",
            "title": "GLM-5.3: Frontier Coding with Emergent Cyber Capabilities",
            "lab": "zai",
            "date": "2026-08-14",
            "printed": "88.8",
            "where": "'Performance across comparison models' table, row 'Terminal Bench 2.1 (Claude Code harness per footnote)', column 'GPT-5.6 Sol'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/9e2ca9dd2373f9f9.png"
      },
      {
        "id": "gpt-5-6-sol:osworld-2-0:offline-partial:1",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "osworld-2-0",
        "benchmark_name": "OSWorld 2.0",
        "setting": "offline-partial",
        "setting_name": "Offline set, partial score",
        "score": 62.6,
        "unit": "%",
        "printed": "62.6%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.6 Sol",
        "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": "computer-use/agentic eval table (HTML), row 'OSWorld 2.0', column 'GPT‑5.6 Sol'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "set, scoring and task release not stated",
        "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/2eabd1ed1ba3ff72.png"
      },
      {
        "id": "gpt-5-6-sol:arc-agi-3:default:1",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "arc-agi-3",
        "benchmark_name": "ARC-AGI-3",
        "setting": "default",
        "setting_name": "",
        "score": 7.78,
        "unit": "%",
        "printed": "7.78%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT‑5.6 Sol",
        "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 eval table (HTML, last table), row 'ARC-AGI-3⁷', column 'GPT‑5.6 Sol'",
        "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-opus-5-system-card",
            "name": "Claude Opus 5 card",
            "title": "Claude Opus 5 System Card",
            "lab": "anthropic",
            "date": "2026-07-24",
            "printed": "7.8",
            "where": "Table 8.1.A (p.153), row ARC-AGI-3, column GPT 5.6 Sol"
          },
          {
            "url": "https://www.anthropic.com/news/claude-opus-5",
            "name": "Claude Opus 5",
            "title": "Introducing Claude Opus 5",
            "lab": "anthropic",
            "date": "2026-07-24",
            "printed": "7.8%",
            "where": "benchmark table image, row ARC-AGI-3, column GPT-5.6 Sol"
          },
          {
            "url": "https://openai.com/index/gpt-6-astra",
            "name": "GPT-6 Astra",
            "title": "GPT-6 Astra: A new generation of intelligence",
            "lab": "openai",
            "date": "2026-09-03",
            "printed": "7.8%",
            "where": "table 8 \"ARC-AGI-3\" row, \"GPT-5.6 Sol\" column"
          },
          {
            "url": "https://openai.com/index/how-two-settings-tripled-our-arc-agi-3-scores",
            "name": "OpenAI post",
            "title": "How enabling two settings tripled our scores on the ARC-AGI-3 benchmark",
            "lab": "openai",
            "date": "2026-07-29",
            "printed": "7.8%",
            "where": "sentence: \"on ARC-AGI-3 ... GPT-5.6 Sol scored just 7.8%\""
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/669da4e40aae1378.png"
      },
      {
        "id": "gpt-5-6-sol:automationbench:default:1",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "automationbench",
        "benchmark_name": "AutomationBench",
        "setting": "default",
        "setting_name": "",
        "score": 18.1,
        "unit": "%",
        "printed": "18.1%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT-5.6 Sol",
        "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-5.6 Sol'",
        "own": true,
        "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-opus-5",
            "name": "Claude Opus 5",
            "title": "Introducing Claude Opus 5",
            "lab": "anthropic",
            "date": "2026-07-24",
            "printed": "18.1%",
            "where": "Benchmark table image, row \"Business workflows AutomationBench\", column \"GPT-5.6 Sol\""
          },
          {
            "url": "https://openai.com/index/gpt-6-astra",
            "name": "GPT-6 Astra",
            "title": "GPT-6 Astra: A new generation of intelligence",
            "lab": "openai",
            "date": "2026-09-03",
            "printed": "18.1%",
            "where": "Professional table, row 'AutomationBench', column 'GPT-5.6 Sol'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/90318aa23c794381.png"
      },
      {
        "id": "gpt-5-6-sol:frontiercode-1-1:main:1",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "frontiercode-1-1",
        "benchmark_name": "FrontierCode 1.1",
        "setting": "main",
        "setting_name": "Main set",
        "score": 47.5,
        "unit": "%",
        "printed": "47.5%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT-5.6 Sol",
        "source": "https://openai.com/index/gpt-6-astra",
        "source_title": "GPT-6 Astra: A new generation of intelligence",
        "source_date": "2026-09-03",
        "where": "'Coding' table, row 'FrontierCode 1.1 Main (score)', column 'GPT-5.6 Sol'",
        "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-opus-5",
            "name": "Claude Opus 5",
            "title": "Introducing Claude Opus 5",
            "lab": "anthropic",
            "date": "2026-07-24",
            "printed": "47.5%",
            "where": "Benchmark table image, row \"Agentic coding FrontierCode v1.1, Main\", column \"GPT-5.6 Sol\""
          },
          {
            "url": "https://www.anthropic.com/claude-opus-5-5",
            "name": "Claude Opus 5.5",
            "title": "Introducing Claude Opus 5.5",
            "lab": "anthropic",
            "date": "2026-09-22",
            "printed": "47.5%",
            "where": "HTML benchmark table, row \"Agentic coding FrontierCode v1.1 (Main)\", column \"GPT-5.6 Sol\""
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/4de8f321fbadf8b8.png"
      },
      {
        "id": "gpt-5-6-sol:frontiercode-1-1:extended:1",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "frontiercode-1-1",
        "benchmark_name": "FrontierCode 1.1",
        "setting": "extended",
        "setting_name": "Extended set",
        "score": 60.6,
        "unit": "%",
        "printed": "60.6%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT-5.6 Sol",
        "source": "https://openai.com/index/gpt-6-astra",
        "source_title": "GPT-6 Astra: A new generation of intelligence",
        "source_date": "2026-09-03",
        "where": "'Coding' table, row 'FrontierCode 1.1 Extended (score)', column 'GPT-5.6 Sol'",
        "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/4de8f321fbadf8b8.png"
      },
      {
        "id": "gpt-5-6-sol:osworld-2-0:offline-partial:2",
        "model": "gpt-5-6-sol",
        "model_name": "GPT-5.6 Sol",
        "model_date": "2026-07-09",
        "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
        "benchmark": "osworld-2-0",
        "benchmark_name": "OSWorld 2.0",
        "setting": "offline-partial",
        "setting_name": "Offline set, partial score",
        "score": 65.7,
        "unit": "%",
        "printed": "65.7%",
        "reported_by": "openai",
        "run_by": null,
        "label": "GPT-5.6 Sol",
        "source": "https://openai.com/index/gpt-6-astra",
        "source_title": "GPT-6 Astra: A new generation of intelligence",
        "source_date": "2026-09-03",
        "where": "'Computer Use' table, row 'OSWorld 2.0 (v2026.08.08, offline set, partial score)', column 'GPT-5.6 Sol2'",
        "own": false,
        "primary": false,
        "parallel": false,
        "notes": "task release v2026.08.08",
        "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/fbd1c3587f0ca67d.png"
      }
    ]
  }
}