{
  "timestamp": "2026-03-12T09:40:09.623487",
  "model_name": "openai-gpt-oss-120b",
  "results": [
    {
      "config": {
        "input_tokens": 50000,
        "output_tokens": 512,
        "batch_size": 64,
        "num_batches": 1,
        "total_requests": 64,
        "actual_input_tokens": 40613
      },
      "success_metrics": {
        "success_rate": 26.56,
        "successful_requests": 17,
        "failed_requests": 47
      },
      "latency": {
        "mean": 101.447,
        "std": 0.463,
        "min": 100.168,
        "max": 102.338,
        "p50": 101.438,
        "p95": 102.129,
        "p99": 102.296,
        "ci_95_lower": 101.227,
        "ci_95_upper": 101.668
      },
      "ttft": {
        "mean": 82.918,
        "std": 6.278,
        "p50": 82.795,
        "p90": 86.47
      },
      "tokens": {
        "total_generated": 8704,
        "content_tokens": 6245,
        "reasoning_tokens": 2459,
        "avg_per_request": 512.0
      },
      "throughput": {
        "concurrent_total_tps": 84.04,
        "concurrent_content_tps": 60.3,
        "requests_per_second": 0.16,
        "actual_wall_time": 103.567,
        "efficiency_percent": 26.02
      },
      "batch_metrics": {
        "num_batches": 1,
        "avg_batch_size": 17.0,
        "avg_batch_throughput": 84.04,
        "min_batch_throughput": 84.04,
        "max_batch_throughput": 84.04
      }
    }
  ]
}