{
  "value": 79.2,
  "display": "79.2",
  "unit": "% resolved",
  "top_entry": "live-SWE-agent + Claude 4.5 Opus medium (20251101)",
  "top_date": "2025-12-15",
  "delta": 0.0,
  "delta_label": "TIED",
  "n_entries": 180,
  "raw": {
    "top_10": [
      {
        "name": "live-SWE-agent + Claude 4.5 Opus medium (20251101)",
        "score": 79.2,
        "date": "2025-12-15",
        "os_model": false,
        "os_system": true
      },
      {
        "name": "Sonar Foundation Agent + Claude 4.5 Opus",
        "score": 79.2,
        "date": "2025-12-05",
        "os_model": false,
        "os_system": false
      },
      {
        "name": "TRAE + Doubao-Seed-Code",
        "score": 78.8,
        "date": "2025-09-28",
        "os_model": false,
        "os_system": true
      },
      {
        "name": "live-SWE-agent + Gemini 3 Pro Preview (2025-11-18)",
        "score": 77.4,
        "date": "2025-11-20",
        "os_model": false,
        "os_system": true
      },
      {
        "name": "Atlassian Rovo Dev (2025-09-02)",
        "score": 76.8,
        "date": "2025-09-02",
        "os_model": false,
        "os_system": false
      },
      {
        "name": "EPAM AI/Run Developer Agent v20250719 + Claude 4 Sonnet",
        "score": 76.8,
        "date": "2025-08-04",
        "os_model": false,
        "os_system": false
      },
      {
        "name": "mini-SWE-agent + Claude 4.5 Opus (high reasoning)",
        "score": 76.8,
        "date": "2026-02-17",
        "os_model": false,
        "os_system": true
      },
      {
        "name": "ACoder",
        "score": 76.4,
        "date": "2025-08-19",
        "os_model": false,
        "os_system": false
      },
      {
        "name": "mini-SWE-agent + Gemini 3 Flash (high reasoning)",
        "score": 75.8,
        "date": "2026-02-17",
        "os_model": false,
        "os_system": true
      },
      {
        "name": "mini-SWE-agent + MiniMax M2.5 (high reasoning)",
        "score": 75.8,
        "date": "2026-02-17",
        "os_model": false,
        "os_system": true
      }
    ],
    "source_url": "https://raw.githubusercontent.com/SWE-bench/swe-bench.github.io/master/data/leaderboards.json",
    "leaderboard_url": "https://www.swebench.com/"
  },
  "status": "live",
  "signal_id": "09",
  "timestamp": "2026-05-26T18:02:57.232356+00:00",
  "last_successful_fetch": "2026-05-26T18:02:57.232356+00:00",
  "stale_at": null,
  "source_url": "https://www.swebench.com/",
  "methodology_version": "v1.0",
  "measurement_risk": "LOW",
  "evidence_strength": "STRONG",
  "why_matters": "Tracks the top published score on SWE-bench Verified (human-validated real GitHub issues) as a proxy for autonomous coding ability. The score reflects a full system — model plus agent scaffolding — is Python/OSS-only, and often trails labs' own marketing claims.",
  "is_estimate": false,
  "is_extrapolated": false,
  "is_annualized": false,
  "snapshot_hash_full": "sha256:c1a75a4dd1bad6127ecf98a91964173e29fea9e02a4577ea842cb35d19d6e01a",
  "snapshot_hash_short": "c1a75a4d"
}