{
  "api_version": "1",
  "updated": "2026-10-26",
  "latest_event": "2026-10-24",
  "rights": "© Frontier Log, all rights reserved.",
  "model": {
    "slug": "qwen3-max-thinking",
    "name": "Qwen3-Max-Thinking",
    "lab": "alibaba",
    "date": "2026-01-26",
    "flagship": false,
    "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
    "releases": [],
    "variant_of": null,
    "made_by": "Alibaba",
    "url": "https://frontierlog.ai/model/qwen3-max-thinking",
    "results": [
      {
        "id": "qwen3-max-thinking:mmlu-pro:default:1",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "mmlu-pro",
        "benchmark_name": "MMLU-Pro",
        "setting": "default",
        "setting_name": "",
        "score": 85.7,
        "unit": "%",
        "printed": "85.7",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking",
        "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
        "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
        "source_date": "2026-01-26",
        "where": "evaluation table, row 'MMLU-Pro', column 'Qwen3-Max-Thinking'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/aed8db26403e0575.png"
      },
      {
        "id": "qwen3-max-thinking:gpqa-diamond:default:1",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 87.4,
        "unit": "%",
        "printed": "87.4",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking",
        "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
        "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
        "source_date": "2026-01-26",
        "where": "evaluation table, row 'GPQA', column 'Qwen3-Max-Thinking'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "without test-time scaling",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/aed8db26403e0575.png"
      },
      {
        "id": "qwen3-max-thinking:hle:no-tools:1",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 30.2,
        "unit": "%",
        "printed": "30.2",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking",
        "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
        "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
        "source_date": "2026-01-26",
        "where": "evaluation table, row 'HLE', column 'Qwen3-Max-Thinking'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "text-only subset",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/aed8db26403e0575.png"
      },
      {
        "id": "qwen3-max-thinking:livecodebench:default:1",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "livecodebench",
        "benchmark_name": "LiveCodeBench",
        "setting": "default",
        "setting_name": "",
        "score": 85.9,
        "unit": "%",
        "printed": "85.9",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking",
        "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
        "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
        "source_date": "2026-01-26",
        "where": "evaluation table, row 'LiveCodeBench v6', column 'Qwen3-Max-Thinking'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "LiveCodeBench v6, problems from Feb to May 2025",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/aed8db26403e0575.png"
      },
      {
        "id": "qwen3-max-thinking:hmmt-2025:no-tools:1",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "hmmt-2025",
        "benchmark_name": "HMMT February 2025",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 98.0,
        "unit": "%",
        "printed": "98.0",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking",
        "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
        "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
        "source_date": "2026-01-26",
        "where": "evaluation table, row 'HMMT Feb 25', column 'Qwen3-Max-Thinking'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/aed8db26403e0575.png"
      },
      {
        "id": "qwen3-max-thinking:swe-bench-verified:default:1",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "swe-bench-verified",
        "benchmark_name": "SWE-bench Verified",
        "setting": "default",
        "setting_name": "",
        "score": 75.3,
        "unit": "%",
        "printed": "75.3",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking",
        "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
        "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
        "source_date": "2026-01-26",
        "where": "evaluation table, row 'SWE Verified', column 'Qwen3-Max-Thinking'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/aed8db26403e0575.png"
      },
      {
        "id": "qwen3-max-thinking:hle:with-tools:1",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "with-tools",
        "setting_name": "With tools",
        "score": 49.8,
        "unit": "%",
        "printed": "49.8",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking",
        "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
        "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
        "source_date": "2026-01-26",
        "where": "evaluation table, row 'HLE (w/ tools)', column 'Qwen3-Max-Thinking'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "text-only subset; with search, HLE-related websites blocked",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/aed8db26403e0575.png"
      },
      {
        "id": "qwen3-max-thinking:gpqa-diamond:default:2",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 92.8,
        "unit": "%",
        "printed": "92.8",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking (with TTS)",
        "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
        "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
        "source_date": "2026-01-26",
        "where": "score chart image (seven panels, below the banner), panel 'GPQA Diamond', first bar, upper value label (legend 'Qwen3-Max-Thinking (with TTS)')",
        "own": true,
        "primary": false,
        "parallel": true,
        "notes": "heavy mode: multi-round test-time scaling over parallel trajectories",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_image": "https://frontierlog.ai/sources/80d8b6e334ec18d8.png",
        "source_image_original": "https://qianwen-res.oss-accelerate-overseas.aliyuncs.com/Qwen3-Max-Thinking/score.png"
      },
      {
        "id": "qwen3-max-thinking:livecodebench:default:2",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "livecodebench",
        "benchmark_name": "LiveCodeBench",
        "setting": "default",
        "setting_name": "",
        "score": 91.4,
        "unit": "%",
        "printed": "91.4",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking (with TTS)",
        "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
        "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
        "source_date": "2026-01-26",
        "where": "score chart image (seven panels, below the banner), panel 'LiveCodeBench (2025.02-2025.05)', first bar, upper value label (legend 'Qwen3-Max-Thinking (with TTS)')",
        "own": true,
        "primary": false,
        "parallel": true,
        "notes": "heavy mode: test-time scaling over parallel trajectories",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_image": "https://frontierlog.ai/sources/80d8b6e334ec18d8.png",
        "source_image_original": "https://qianwen-res.oss-accelerate-overseas.aliyuncs.com/Qwen3-Max-Thinking/score.png"
      },
      {
        "id": "qwen3-max-thinking:hle:no-tools:2",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 36.5,
        "unit": "%",
        "printed": "36.5",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking (with TTS)",
        "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
        "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
        "source_date": "2026-01-26",
        "where": "score chart image (seven panels, below the banner), panel 'Humanity's Last Exam', first bar, upper value label (legend 'Qwen3-Max-Thinking (with TTS)')",
        "own": true,
        "primary": false,
        "parallel": true,
        "notes": "heavy mode: test-time scaling over parallel trajectories; text-only subset",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_image": "https://frontierlog.ai/sources/80d8b6e334ec18d8.png",
        "source_image_original": "https://qianwen-res.oss-accelerate-overseas.aliyuncs.com/Qwen3-Max-Thinking/score.png"
      },
      {
        "id": "qwen3-max-thinking:hle:with-tools:2",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "with-tools",
        "setting_name": "With tools",
        "score": 58.3,
        "unit": "%",
        "printed": "58.3",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking (with TTS)",
        "source": "https://qwen.ai/blog?id=qwen3-max-thinking",
        "source_title": "Pushing Qwen3-Max-Thinking Beyond its Limits",
        "source_date": "2026-01-26",
        "where": "score chart image (seven panels, below the banner), panel 'Humanity's Last Exam (with Search)', first bar, upper value label (legend 'Qwen3-Max-Thinking (with TTS)')",
        "own": true,
        "primary": false,
        "parallel": true,
        "notes": "heavy mode: test-time scaling over parallel trajectories; with search",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_image": "https://frontierlog.ai/sources/80d8b6e334ec18d8.png",
        "source_image_original": "https://qianwen-res.oss-accelerate-overseas.aliyuncs.com/Qwen3-Max-Thinking/score.png"
      },
      {
        "id": "qwen3-max-thinking:browsecomp:default:1",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "browsecomp",
        "benchmark_name": "BrowseComp",
        "setting": "default",
        "setting_name": "",
        "score": 53.9,
        "unit": "%",
        "printed": "53.9",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "Qwen3.5 post, 'Performance' section, 'Language' table, row 'BrowseComp', column 'Qwen3-Max-Thinking'",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "search agent; context folding at 256K tokens",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/a2e98da83152d308.png"
      },
      {
        "id": "qwen3-max-thinking:terminal-bench-2-0:default:1",
        "model": "qwen3-max-thinking",
        "model_name": "Qwen3-Max-Thinking",
        "model_date": "2026-01-26",
        "event": "https://frontierlog.ai/event/2026-01-26-qwen3-max-thinking",
        "benchmark": "terminal-bench-2-0",
        "benchmark_name": "Terminal-Bench 2.0",
        "setting": "default",
        "setting_name": "",
        "score": 22.5,
        "unit": "%",
        "printed": "22.5",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3-Max-Thinking",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "Qwen3.5 post, 'Performance' section, 'Language' table, row 'Terminal Bench 2', column 'Qwen3-Max-Thinking'",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/a2e98da83152d308.png"
      }
    ]
  }
}