{
  "api_version": "1",
  "updated": "2026-10-28",
  "latest_event": "2026-10-27",
  "rights": "© Frontier Log, all rights reserved.",
  "model": {
    "slug": "glm-4-5-air",
    "name": "GLM-4.5-Air",
    "lab": "zai",
    "date": "2025-07-28",
    "source": "https://z.ai/blog/glm-4.5",
    "source_title": "GLM-4.5: Reasoning, Coding, and Agentic Abililties",
    "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-air",
    "results": [
      {
        "id": "glm-4-5-air:mmlu-pro:default:1",
        "model": "glm-4-5-air",
        "model_name": "GLM-4.5-Air",
        "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": 81.4,
        "unit": "%",
        "printed": "81.4",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5-Air",
        "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 table (HTML), row 'MMLU Pro', column 'GLM-4.5-Air'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "thinking mode",
        "caveat": "",
        "human_verified": false,
        "human_verified_on": null,
        "check": "double_entry",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": []
      },
      {
        "id": "glm-4-5-air:aime-2024:no-tools:1",
        "model": "glm-4-5-air",
        "model_name": "GLM-4.5-Air",
        "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": 89.4,
        "unit": "%",
        "printed": "89.4",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5-Air",
        "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 table (HTML), row 'AIME24', column 'GLM-4.5-Air'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "thinking mode; average over 32 samples, answers checked by an LLM",
        "caveat": "",
        "human_verified": false,
        "human_verified_on": null,
        "check": "double_entry",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": []
      },
      {
        "id": "glm-4-5-air:livecodebench:default:1",
        "model": "glm-4-5-air",
        "model_name": "GLM-4.5-Air",
        "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": 70.7,
        "unit": "%",
        "printed": "70.7",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5-Air",
        "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 table (HTML), row 'LiveCodeBench 2407-2501', column 'GLM-4.5-Air'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "thinking mode; problems from Jul 2024 to Jan 2025",
        "caveat": "",
        "human_verified": false,
        "human_verified_on": null,
        "check": "double_entry",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": []
      },
      {
        "id": "glm-4-5-air:swe-bench-verified:default:1",
        "model": "glm-4-5-air",
        "model_name": "GLM-4.5-Air",
        "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": 57.6,
        "unit": "%",
        "printed": "57.6",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5-Air",
        "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 table (HTML), row 'SWE-bench Verified', column 'GLM-4.5-Air'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "OpenHands v0.34.0; up to 100 iterations; 128K context with history truncation",
        "caveat": "",
        "human_verified": false,
        "human_verified_on": null,
        "check": "double_entry",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": []
      },
      {
        "id": "glm-4-5-air:terminal-bench-1-0:default:1",
        "model": "glm-4-5-air",
        "model_name": "GLM-4.5-Air",
        "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": 30.0,
        "unit": "%",
        "printed": "30.0",
        "reported_by": "zai",
        "run_by": null,
        "label": "GLM-4.5-Air",
        "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 table (HTML), row 'SWE-Terminal-Bench' (footnote: Terminal-Bench with Terminus), column 'GLM-4.5-Air'",
        "own": true,
        "primary": true,
        "parallel": false,
        "notes": "Terminus harness, standard function calling",
        "caveat": "",
        "human_verified": false,
        "human_verified_on": null,
        "check": "double_entry",
        "corroborated": false,
        "corroborated_on": null,
        "corroborated_by": []
      }
    ]
  }
}