{
 "@cf/zai-org/glm-5.2": {
  "n": 118,
  "valid": 118,
  "accuracy": 0.89,
  "macro_f1": 0.91,
  "per_label": {
   "aware": {
    "p": 1.0,
    "r": 1.0,
    "f1": 1.0
   },
   "substantive": {
    "p": 0.76,
    "r": 0.97,
    "f1": 0.86
   },
   "perfunctory": {
    "p": 0.97,
    "r": 0.78,
    "f1": 0.87
   }
  },
  "aware_caught": "27/27",
  "aware_false_alarms": 0,
  "quote_ok": 0.98,
  "p50_ms": 1078,
  "usd_per_1k_items": 0.614,
  "errors": []
 },
 "@cf/openai/gpt-oss-120b": {
  "n": 118,
  "valid": 115,
  "accuracy": 0.864,
  "macro_f1": 0.89,
  "per_label": {
   "aware": {
    "p": 1.0,
    "r": 0.93,
    "f1": 0.96
   },
   "substantive": {
    "p": 0.75,
    "r": 1.0,
    "f1": 0.86
   },
   "perfunctory": {
    "p": 1.0,
    "r": 0.74,
    "f1": 0.85
   }
  },
  "aware_caught": "25/27",
  "aware_false_alarms": 0,
  "quote_ok": 0.69,
  "p50_ms": 4910,
  "usd_per_1k_items": 0.5132,
  "errors": [
   [
    "truncated: length",
    3
   ]
  ]
 },
 "@cf/cloudflare/clef": {
  "n": 118,
  "valid": 118,
  "accuracy": 0.907,
  "macro_f1": 0.917,
  "per_label": {
   "aware": {
    "p": 1.0,
    "r": 0.89,
    "f1": 0.94
   },
   "substantive": {
    "p": 0.83,
    "r": 0.97,
    "f1": 0.9
   },
   "perfunctory": {
    "p": 0.94,
    "r": 0.88,
    "f1": 0.91
   }
  },
  "aware_caught": "24/27",
  "aware_false_alarms": 0,
  "quote_ok": null,
  "p50_ms": 524,
  "usd_per_1k_items": 0.184,
  "errors": []
 },
 "typesafe/jev": {
  "n": 118,
  "valid": 118,
  "accuracy": 0.932,
  "macro_f1": 0.94,
  "per_label": {
   "aware": {
    "p": 1.0,
    "r": 0.93,
    "f1": 0.96
   },
   "substantive": {
    "p": 0.88,
    "r": 0.95,
    "f1": 0.92
   },
   "perfunctory": {
    "p": 0.94,
    "r": 0.94,
    "f1": 0.94
   }
  },
  "aware_caught": "25/27",
  "aware_false_alarms": 0,
  "quote_ok": null,
  "p50_ms": 692,
  "usd_per_1k_items": 0.0405,
  "errors": []
 },
 "@cf/moonshotai/kimi-k2.6": {
  "n": 118,
  "valid": 118,
  "accuracy": 0.898,
  "macro_f1": 0.913,
  "per_label": {
   "aware": {
    "p": 1.0,
    "r": 1.0,
    "f1": 1.0
   },
   "substantive": {
    "p": 0.85,
    "r": 0.85,
    "f1": 0.85
   },
   "perfunctory": {
    "p": 0.88,
    "r": 0.9,
    "f1": 0.89
   }
  },
  "aware_caught": "27/27",
  "aware_false_alarms": 0,
  "quote_ok": 0.97,
  "p50_ms": 2299,
  "usd_per_1k_items": 0.3363,
  "errors": []
 },
 "@cf/google/gemma-4-26b-a4b-it": {
  "n": 118,
  "valid": 118,
  "accuracy": 0.915,
  "macro_f1": 0.923,
  "per_label": {
   "aware": {
    "p": 1.0,
    "r": 0.93,
    "f1": 0.96
   },
   "substantive": {
    "p": 0.88,
    "r": 0.93,
    "f1": 0.9
   },
   "perfunctory": {
    "p": 0.9,
    "r": 0.92,
    "f1": 0.91
   }
  },
  "aware_caught": "25/27",
  "aware_false_alarms": 0,
  "quote_ok": 0.92,
  "p50_ms": 1200,
  "usd_per_1k_items": 0.075,
  "errors": []
 },
 "@cf/nvidia/nemotron-3-120b-a12b": {
  "n": 118,
  "valid": 118,
  "accuracy": 0.864,
  "macro_f1": 0.887,
  "per_label": {
   "aware": {
    "p": 1.0,
    "r": 0.85,
    "f1": 0.92
   },
   "substantive": {
    "p": 0.89,
    "r": 0.82,
    "f1": 0.86
   },
   "perfunctory": {
    "p": 0.84,
    "r": 0.92,
    "f1": 0.88
   }
  },
  "aware_caught": "23/27",
  "aware_false_alarms": 0,
  "quote_ok": 0.98,
  "p50_ms": 1356,
  "usd_per_1k_items": 0.4079,
  "errors": []
 },
 "@cf/deepseek-ai/deepseek-v4-pro-0813": {
  "n": 118,
  "valid": 118,
  "accuracy": 0.873,
  "macro_f1": 0.88,
  "per_label": {
   "aware": {
    "p": 1.0,
    "r": 0.89,
    "f1": 0.94
   },
   "substantive": {
    "p": 0.91,
    "r": 0.75,
    "f1": 0.82
   },
   "perfunctory": {
    "p": 0.8,
    "r": 0.98,
    "f1": 0.88
   }
  },
  "aware_caught": "24/27",
  "aware_false_alarms": 0,
  "quote_ok": 1.0,
  "p50_ms": 2476,
  "usd_per_1k_items": 0.6288,
  "errors": []
 },
 "_note": "Gold set: 73 random dev citations + 45 retraction-aware challenge citations, labeled by Claude (codebook v2). Clef priced at $0.24/M input tokens, Jev at $0.042/M, others from Workers AI neurons. 2026-10-13."
}