{
  "aliases": [],
  "authors": [],
  "details": {
    "benchmark_protocol": {
      "actions": [],
      "counts": [
        {
          "basis": "reported",
          "evidence_url": "https://raw.githubusercontent.com/open-compass/MMBench-GUI/HEAD/README.md",
          "population": "reported multi-level benchmark",
          "qualifier": "more-than",
          "unit": "tasks",
          "value": 8000
        }
      ],
      "evaluation": "Level-specific evaluation; Efficiency–Quality Area (EQA) combines navigation quality and efficiency.",
      "evidence_urls": [
        "https://raw.githubusercontent.com/open-compass/MMBench-GUI/HEAD/README.md"
      ],
      "focus": [
        "screen understanding",
        "grounding",
        "task automation",
        "task collaboration"
      ],
      "notes": "The README documents four levels, but its release notes and TODOs distinguish released L1/L2 evaluation from pending L3/L4 integration. Do not treat every level as the same evaluator or a complete implemented interface.",
      "observations": [
        "screenshots"
      ],
      "setting": "mixed"
    }
  },
  "environments": [
    "cross-platform"
  ],
  "evidence": [
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-10-16T06:35:58Z",
      "role": "primary",
      "sha256": "35db9fca3ed1c1a3513924fefff4ccd224bc70cd7eb4bf8c54fcb9048851988e",
      "supports": [
        "name",
        "kind",
        "summary",
        "environments",
        "topics",
        "scope",
        "relations",
        "details.benchmark_protocol"
      ],
      "url": "https://raw.githubusercontent.com/open-compass/MMBench-GUI/HEAD/README.md"
    }
  ],
  "id": "github:open-compass/mmbench-gui",
  "identifiers": {},
  "kind": "benchmark",
  "last_verified_at": "2026-10-16T06:35:58Z",
  "licenses": {
    "code": null,
    "data": null,
    "document": null,
    "weights": null
  },
  "links": [
    {
      "rel": "primary",
      "url": "https://github.com/open-compass/MMBench-GUI"
    }
  ],
  "name": "MMBench-GUI",
  "notes": "",
  "published_at": null,
  "relations": [
    {
      "evidence_url": "https://raw.githubusercontent.com/open-compass/MMBench-GUI/HEAD/README.md",
      "target_id": "arxiv:2507.19478",
      "type": "documented_by"
    }
  ],
  "review_status": "curated",
  "scope": "core",
  "summary": "Four-level GUI-agent benchmark across six platforms, from content understanding and element grounding to task automation and collaboration, scored with an efficiency-quality area metric.",
  "topics": [
    "evaluation",
    "grounding"
  ],
  "updated_at": null,
  "url": "https://github.com/open-compass/MMBench-GUI"
}