{
  "coding": {
    "model": "qwen2.5-coder:1.5b",
    "score_pct": 100.0,
    "tasks_passed": 10,
    "tasks_total": 10,
    "ci_95_pct": [
      72.2,
      100.0
    ],
    "breakdown": {
      "algorithms": "4/4 (Fibonacci, Palindrome, Merge Intervals, Binary Search)",
      "bug_fixes": "3/3 (Loop Bound, KeyError Guard, ZeroDivision & Rounding)",
      "sql_queries": "3/3 (JOIN & GROUP BY, HAVING Aggregate, Date Filtering)"
    }
  },
  "structured_extraction": {
    "model": "qwen2.5-coder:1.5b",
    "score_pct": 90.0,
    "tasks_passed": 9,
    "tasks_total": 10,
    "ci_95_pct": [
      59.6,
      98.2
    ],
    "breakdown": {
      "json_schema": "4/4 valid types and required properties",
      "negative_controls": "2/2 zero hallucinated extra keys",
      "multilingual_hi_en": "3/4 (Failed Hindi phonetic transliteration 'Jaya Pur' vs 'Jaipur')"
    }
  },
  "retrieval_offline_rag": {
    "model": "nomic-embed-text (137M nomic-bert)",
    "recall_at_1_pct": 100.0,
    "recall_at_5_pct": 100.0,
    "mrr": 1.0,
    "avg_latency_ms": 3028.17,
    "vram_used_mb": 323.0,
    "embedding_dim": 768
  },
  "reasoning_v2": {
    "model": "qwen2.5-coder:1.5b",
    "total_tasks": 30,
    "hotpotqa": {
      "score_pct": 93.3,
      "ci_95_pct": [
        70.2,
        98.8
      ],
      "passed": 14,
      "total": 15
    },
    "alfworld": {
      "score_pct": 46.7,
      "ci_95_pct": [
        24.8,
        69.9
      ],
      "passed": 7,
      "total": 15
    }
  }
}