{
  "api_version": "1",
  "updated": "2026-10-26",
  "latest_event": "2026-10-24",
  "rights": "© Frontier Log, all rights reserved.",
  "model": {
    "slug": "glm-4-5",
    "name": "GLM-4.5",
    "lab": "zai",
    "date": "2025-07-28",
    "flagship": false,
    "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
    "releases": [],
    "variant_of": null,
    "made_by": "Z.ai",
    "url": "https://frontierlog.ai/model/glm-4-5",
    "results": [
      {
        "id": "glm-4-5:browsecomp:default:1",
        "model": "glm-4-5",
        "model_name": "GLM-4.5",
        "model_date": "2025-07-28",
        "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
        "benchmark": "browsecomp",
        "benchmark_name": "BrowseComp",
        "setting": "default",
        "setting_name": "",
        "score": 26.4,
        "unit": "%",
        "printed": "26.4",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5",
        "source": "https://z.ai/blog/glm-4.5",
        "source_title": "GLM-4.5: Reasoning, Coding, and Agentic Abililties",
        "source_date": "2025-07-28",
        "where": "'Agentic Tasks' section, results table, row 'BrowseComp', column 'GLM-4.5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "web browsing tool",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://z.ai/blog/glm-4.6",
            "name": "GLM-4.6",
            "title": "GLM-4.6: Advanced Agentic, Reasoning and Coding Capabilities",
            "lab": "zai",
            "date": "2025-09-30",
            "printed": "26.4",
            "where": "Chart 'LLM Performance Evaluation: Agentic, Reasoning and Coding', panel 'BrowseComp', bar 'GLM-4.5'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/ab05b437e471e43a.png"
      },
      {
        "id": "glm-4-5:mmlu-pro:default:1",
        "model": "glm-4-5",
        "model_name": "GLM-4.5",
        "model_date": "2025-07-28",
        "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
        "benchmark": "mmlu-pro",
        "benchmark_name": "MMLU-Pro",
        "setting": "default",
        "setting_name": "",
        "score": 84.6,
        "unit": "%",
        "printed": "84.6",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5",
        "source": "https://z.ai/blog/glm-4.5",
        "source_title": "GLM-4.5: Reasoning, Coding, and Agentic Abililties",
        "source_date": "2025-07-28",
        "where": "'Reasoning' section table, row 'MMLU Pro', column 'GLM-4.5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "thinking mode",
        "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/9690be94d5b435fb.png"
      },
      {
        "id": "glm-4-5:aime-2024:no-tools:1",
        "model": "glm-4-5",
        "model_name": "GLM-4.5",
        "model_date": "2025-07-28",
        "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
        "benchmark": "aime-2024",
        "benchmark_name": "AIME 2024",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 91.0,
        "unit": "%",
        "printed": "91.0",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5",
        "source": "https://z.ai/blog/glm-4.5",
        "source_title": "GLM-4.5: Reasoning, Coding, and Agentic Abililties",
        "source_date": "2025-07-28",
        "where": "'Reasoning' section table, row 'AIME24', column 'GLM-4.5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "thinking mode; average over 32 samples, answers checked by an LLM",
        "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/9690be94d5b435fb.png"
      },
      {
        "id": "glm-4-5:gpqa-diamond:default:1",
        "model": "glm-4-5",
        "model_name": "GLM-4.5",
        "model_date": "2025-07-28",
        "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 79.1,
        "unit": "%",
        "printed": "79.1",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5",
        "source": "https://z.ai/blog/glm-4.5",
        "source_title": "GLM-4.5: Reasoning, Coding, and Agentic Abililties",
        "source_date": "2025-07-28",
        "where": "'Reasoning' section table, row 'GPQA', column 'GLM-4.5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "thinking mode; average over 8 samples",
        "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/9690be94d5b435fb.png"
      },
      {
        "id": "glm-4-5:hle:no-tools:1",
        "model": "glm-4-5",
        "model_name": "GLM-4.5",
        "model_date": "2025-07-28",
        "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
        "benchmark": "hle",
        "benchmark_name": "Humanity's Last Exam",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 14.4,
        "unit": "%",
        "printed": "14.4",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5",
        "source": "https://z.ai/blog/glm-4.5",
        "source_title": "GLM-4.5: Reasoning, Coding, and Agentic Abililties",
        "source_date": "2025-07-28",
        "where": "'Reasoning' section table, row 'HLE', column 'GLM-4.5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "thinking mode; text-only questions; GPT-4o judge",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://z.ai/blog/glm-4.6",
            "name": "GLM-4.6",
            "title": "GLM-4.6: Advanced Agentic, Reasoning and Coding Capabilities",
            "lab": "zai",
            "date": "2025-09-30",
            "printed": "14.4",
            "where": "Chart 'LLM Performance Evaluation: Agentic, Reasoning and Coding', panel 'HLE', bar 'GLM-4.5' (base label; only GLM-4.6's separate value is marked 'w/ Tools')"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/9690be94d5b435fb.png"
      },
      {
        "id": "glm-4-5:livecodebench:default:1",
        "model": "glm-4-5",
        "model_name": "GLM-4.5",
        "model_date": "2025-07-28",
        "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
        "benchmark": "livecodebench",
        "benchmark_name": "LiveCodeBench",
        "setting": "default",
        "setting_name": "",
        "score": 72.9,
        "unit": "%",
        "printed": "72.9",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5",
        "source": "https://z.ai/blog/glm-4.5",
        "source_title": "GLM-4.5: Reasoning, Coding, and Agentic Abililties",
        "source_date": "2025-07-28",
        "where": "'Reasoning' section table, row 'LiveCodeBench 2407-2501', column 'GLM-4.5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "thinking mode; problems from Jul 2024 to Jan 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/9690be94d5b435fb.png"
      },
      {
        "id": "glm-4-5:swe-bench-verified:default:1",
        "model": "glm-4-5",
        "model_name": "GLM-4.5",
        "model_date": "2025-07-28",
        "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
        "benchmark": "swe-bench-verified",
        "benchmark_name": "SWE-bench Verified",
        "setting": "default",
        "setting_name": "",
        "score": 64.2,
        "unit": "%",
        "printed": "64.2",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5",
        "source": "https://z.ai/blog/glm-4.5",
        "source_title": "GLM-4.5: Reasoning, Coding, and Agentic Abililties",
        "source_date": "2025-07-28",
        "where": "'Coding' section table, row 'SWE-bench Verified', column 'GLM-4.5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "OpenHands v0.34.0; up to 100 iterations; 128K context with history truncation",
        "caveat": "",
        "human_verified": true,
        "human_verified_on": "2026-10-19",
        "check": "human",
        "corroborated": true,
        "corroborated_on": "2026-10-19",
        "corroborated_by": [
          {
            "url": "https://z.ai/blog/glm-4.6",
            "name": "GLM-4.6",
            "title": "GLM-4.6: Advanced Agentic, Reasoning and Coding Capabilities",
            "lab": "zai",
            "date": "2025-09-30",
            "printed": "64.2",
            "where": "Chart 'LLM Performance Evaluation: Agentic, Reasoning and Coding', panel 'SWE-bench Verified', bar 'GLM-4.5'"
          }
        ],
        "source_capture": "https://frontierlog.ai/captures/7c0c7768432b93d8.png"
      },
      {
        "id": "glm-4-5:terminal-bench-1-0:default:1",
        "model": "glm-4-5",
        "model_name": "GLM-4.5",
        "model_date": "2025-07-28",
        "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
        "benchmark": "terminal-bench-1-0",
        "benchmark_name": "Terminal-Bench 1.0",
        "setting": "default",
        "setting_name": "",
        "score": 37.5,
        "unit": "%",
        "printed": "37.5",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5",
        "source": "https://z.ai/blog/glm-4.5",
        "source_title": "GLM-4.5: Reasoning, Coding, and Agentic Abililties",
        "source_date": "2025-07-28",
        "where": "'Coding' section table, row 'SWE-Terminal-Bench', column 'GLM-4.5'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "Terminus harness, standard function calling",
        "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/7c0c7768432b93d8.png"
      },
      {
        "id": "glm-4-5:aime-2025:no-tools:1",
        "model": "glm-4-5",
        "model_name": "GLM-4.5",
        "model_date": "2025-07-28",
        "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
        "benchmark": "aime-2025",
        "benchmark_name": "AIME 2025",
        "setting": "no-tools",
        "setting_name": "No tools",
        "score": 85.4,
        "unit": "%",
        "printed": "85.4",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5",
        "source": "https://z.ai/blog/glm-4.6",
        "source_title": "GLM-4.6: Advanced Agentic, Reasoning and Coding Capabilities",
        "source_date": "2025-09-30",
        "where": "GLM-4.6 post, benchmark chart image (eight panels), panel 'AIME 25', second bar 'GLM-4.5', value label",
        "own": false,
        "primary": true,
        "parallel": false,
        "notes": "128K context",
        "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/e2ec892bfba7273b.png",
        "source_image_original": "https://z-cdn.chatglm.cn/z-blog/glm-4-6/coding_benchmark.png"
      },
      {
        "id": "glm-4-5:gpqa-diamond:default:2",
        "model": "glm-4-5",
        "model_name": "GLM-4.5",
        "model_date": "2025-07-28",
        "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
        "benchmark": "gpqa-diamond",
        "benchmark_name": "GPQA Diamond",
        "setting": "default",
        "setting_name": "",
        "score": 79.9,
        "unit": "%",
        "printed": "79.9",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5",
        "source": "https://z.ai/blog/glm-4.6",
        "source_title": "GLM-4.6: Advanced Agentic, Reasoning and Coding Capabilities",
        "source_date": "2025-09-30",
        "where": "GLM-4.6 post, benchmark chart image (eight panels), panel 'GPQA', second bar 'GLM-4.5', value label",
        "own": false,
        "primary": false,
        "parallel": false,
        "notes": "128K context",
        "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/e2ec892bfba7273b.png",
        "source_image_original": "https://z-cdn.chatglm.cn/z-blog/glm-4-6/coding_benchmark.png"
      },
      {
        "id": "glm-4-5:livecodebench:default:2",
        "model": "glm-4-5",
        "model_name": "GLM-4.5",
        "model_date": "2025-07-28",
        "event": "https://frontierlog.ai/event/2025-07-28-glm-4-5",
        "benchmark": "livecodebench",
        "benchmark_name": "LiveCodeBench",
        "setting": "default",
        "setting_name": "",
        "score": 63.3,
        "unit": "%",
        "printed": "63.3",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5",
        "source": "https://z.ai/blog/glm-4.6",
        "source_title": "GLM-4.6: Advanced Agentic, Reasoning and Coding Capabilities",
        "source_date": "2025-09-30",
        "where": "GLM-4.6 post, benchmark chart image (eight panels), panel 'LiveCodeBench v6', second bar 'GLM-4.5', value label",
        "own": false,
        "primary": false,
        "parallel": false,
        "notes": "LiveCodeBench v6; 128K context",
        "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/e2ec892bfba7273b.png",
        "source_image_original": "https://z-cdn.chatglm.cn/z-blog/glm-4-6/coding_benchmark.png"
      }
    ]
  }
}