{
  "api_version": "1",
  "updated": "2026-10-26",
  "latest_event": "2026-10-24",
  "rights": "© Frontier Log, all rights reserved.",
  "model": {
    "slug": "qwen3-5-397b-a17b",
    "name": "Qwen3.5-397B-A17B",
    "lab": "alibaba",
    "date": "2026-02-16",
    "source": "https://qwen.ai/blog?id=qwen3.5",
    "source_title": "Qwen3.5: Towards Native Multimodal Agents",
    "flagship": false,
    "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
    "releases": [],
    "variant_of": null,
    "made_by": "Alibaba",
    "url": "https://frontierlog.ai/model/qwen3-5-397b-a17b",
    "results": [
      {
        "id": "qwen3-5-397b-a17b:mmlu-pro:default:1",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "mmlu-pro",
        "benchmark_name": "MMLU-Pro",
        "setting": "default",
        "setting_name": "",
        "score": 87.8,
        "unit": "%",
        "printed": "87.8",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "'Performance' section, 'Language' table, row 'MMLU-Pro', column 'Qwen3.5-397B-A17B'",
        "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/a2e98da83152d308.png"
      },
      {
        "id": "qwen3-5-397b-a17b:gpqa-diamond:default:1",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 88.4,
        "unit": "%",
        "printed": "88.4",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "'Performance' section, 'Language' table, row 'GPQA', column 'Qwen3.5-397B-A17B'",
        "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/a2e98da83152d308.png"
      },
      {
        "id": "qwen3-5-397b-a17b:hle:no-tools:1",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 28.7,
        "unit": "%",
        "printed": "28.7",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "'Performance' section, 'Language' table, row 'HLE', column 'Qwen3.5-397B-A17B'",
        "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/a2e98da83152d308.png"
      },
      {
        "id": "qwen3-5-397b-a17b:livecodebench:default:1",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "livecodebench",
        "benchmark_name": "LiveCodeBench",
        "setting": "default",
        "setting_name": "",
        "score": 83.6,
        "unit": "%",
        "printed": "83.6",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "'Performance' section, 'Language' table, row 'LiveCodeBench v6', column 'Qwen3.5-397B-A17B'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "LiveCodeBench v6",
        "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": "qwen3-5-397b-a17b:hmmt-2025:no-tools:1",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "hmmt-2025",
        "benchmark_name": "HMMT February 2025",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 94.8,
        "unit": "%",
        "printed": "94.8",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "'Performance' section, 'Language' table, row 'HMMT Feb 25', column 'Qwen3.5-397B-A17B'",
        "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/a2e98da83152d308.png"
      },
      {
        "id": "qwen3-5-397b-a17b:hle:with-tools:1",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "with-tools",
        "setting_name": "With tools",
        "score": 48.3,
        "unit": "%",
        "printed": "48.3",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "'Performance' section, 'Language' table, row 'HLE w/ tool', column 'Qwen3.5-397B-A17B'",
        "own": true,
        "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": "qwen3-5-397b-a17b:browsecomp:default:1",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "browsecomp",
        "benchmark_name": "BrowseComp",
        "setting": "default",
        "setting_name": "",
        "score": 78.6,
        "unit": "%",
        "printed": "78.6",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "'Performance' section, 'Language' table, row 'BrowseComp', column 'Qwen3.5-397B-A17B', second value in the cell (after the slash)",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "discard-all context strategy",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://mistral.ai/news/vibe-remote-agents-mistral-medium-3-5/",
            "name": "Mistral Medium 3.5",
            "title": "Remote agents in Vibe. Powered by Mistral Medium 3.5.",
            "lab": "mistral",
            "date": "2026-05-22",
            "printed": "78.6",
            "where": "Chart 'Agentic Benchmarks vs competing models', group 'BrowseComp****' (footnote: self-reported, Mistral uses context management, discard-all at 100k), bar 'Qwen3.5 397B A17B'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/a2e98da83152d308.png"
      },
      {
        "id": "qwen3-5-397b-a17b:browsecomp:default:2",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "browsecomp",
        "benchmark_name": "BrowseComp",
        "setting": "default",
        "setting_name": "",
        "score": 69.0,
        "unit": "%",
        "printed": "69.0",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "'Performance' section, 'Language' table, row 'BrowseComp', column 'Qwen3.5-397B-A17B', first value in the cell (before the slash)",
        "own": true,
        "primary": false,
        "parallel": false,
        "notes": "simple context folding",
        "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": "qwen3-5-397b-a17b:swe-bench-verified:default:1",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "swe-bench-verified",
        "benchmark_name": "SWE-bench Verified",
        "setting": "default",
        "setting_name": "",
        "score": 76.4,
        "unit": "%",
        "printed": "76.4",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "'Performance' section, 'Language' table, row 'SWE-bench Verified', column 'Qwen3.5-397B-A17B'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://mistral.ai/news/vibe-remote-agents-mistral-medium-3-5/",
            "name": "Mistral Medium 3.5",
            "title": "Remote agents in Vibe. Powered by Mistral Medium 3.5.",
            "lab": "mistral",
            "date": "2026-05-22",
            "printed": "76.4",
            "where": "Chart 'Agentic Benchmarks vs competing models', group 'SWE-Bench verified***' (footnote: self-reported; max reasoning), bar 'Qwen3.5 397B A17B'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/a2e98da83152d308.png"
      },
      {
        "id": "qwen3-5-397b-a17b:terminal-bench-2-0:default:1",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "terminal-bench-2-0",
        "benchmark_name": "Terminal-Bench 2.0",
        "setting": "default",
        "setting_name": "",
        "score": 52.5,
        "unit": "%",
        "printed": "52.5",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "'Performance' section, 'Language' table, row 'Terminal Bench 2', column 'Qwen3.5-397B-A17B'",
        "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/a2e98da83152d308.png"
      },
      {
        "id": "qwen3-5-397b-a17b:osworld-verified:default:1",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "osworld-verified",
        "benchmark_name": "OSWorld-Verified",
        "setting": "default",
        "setting_name": "",
        "score": 62.2,
        "unit": "%",
        "printed": "62.2",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.5",
        "source_title": "Qwen3.5: Towards Native Multimodal Agents",
        "source_date": "2026-02-16",
        "where": "'Vision Language' table, row 'OSWorld-Verified' (Visual Agent section), column 'Qwen3.5-397B-A17B'",
        "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/f828906ff2951f75.png"
      },
      {
        "id": "qwen3-5-397b-a17b:swe-bench-verified:default:2",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "swe-bench-verified",
        "benchmark_name": "SWE-bench Verified",
        "setting": "default",
        "setting_name": "",
        "score": 76.2,
        "unit": "%",
        "printed": "76.2",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.6",
        "source_title": "Qwen3.6-Plus: Towards Real World Agents",
        "source_date": "2026-04-02",
        "where": "Qwen3.6-Plus post, 'Performance' section, 'Language' table, row 'SWE-bench Verified', column 'Qwen3.5-397B-A17B'",
        "own": false,
        "primary": false,
        "parallel": false,
        "notes": "internal harness with bash and file-edit tools; 200K context",
        "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/6f4855b376c670b4.png"
      },
      {
        "id": "qwen3-5-397b-a17b:swe-bench-pro:default:1",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "swe-bench-pro",
        "benchmark_name": "SWE-bench Pro",
        "setting": "default",
        "setting_name": "",
        "score": 50.9,
        "unit": "%",
        "printed": "50.9",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.6",
        "source_title": "Qwen3.6-Plus: Towards Real World Agents",
        "source_date": "2026-04-02",
        "where": "Qwen3.6-Plus post, 'Performance' section, 'Language' table, row 'SWE-bench Pro', column 'Qwen3.5-397B-A17B'",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "internal harness with bash and file-edit tools; 200K context",
        "caveat": "Run on the lab's own corrected version of the public set",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": [],
        "source_capture": "https://frontierlog.ai/captures/6f4855b376c670b4.png"
      },
      {
        "id": "qwen3-5-397b-a17b:mcp-atlas:default:1",
        "model": "qwen3-5-397b-a17b",
        "model_name": "Qwen3.5-397B-A17B",
        "model_date": "2026-02-16",
        "event": "https://frontierlog.ai/event/2026-02-16-qwen3-5",
        "benchmark": "mcp-atlas",
        "benchmark_name": "MCP Atlas",
        "setting": "default",
        "setting_name": "",
        "score": 74.2,
        "unit": "%",
        "printed": "74.2",
        "reported_by": "alibaba",
        "run_by": null,
        "label": "Qwen3.5-397B-A17B",
        "source": "https://qwen.ai/blog?id=qwen3.6",
        "source_title": "Qwen3.6-Plus: Towards Real World Agents",
        "source_date": "2026-04-02",
        "where": "Qwen3.6-Plus post, 'Performance' section, 'Language' table, row 'MCP-Atlas', column 'Qwen3.5-397B-A17B'",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "public set; Gemini 2.5 Pro judge",
        "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/6f4855b376c670b4.png"
      }
    ]
  }
}