metrics/inference_benchmark.json

{
  "stage": 3,
  "process": "fresh_process_cpu_inference_benchmark",
  "device": "cpu",
  "batch_size": 1,
  "warmup_batches": 5,
  "measured_batches": 100,
  "measured_sequences": 100,
  "measured_tokens_including_special": 9362,
  "elapsed_seconds": 5.222225,
  "sequences_per_second": 19.148927,
  "tokens_per_second": 1792.722588,
  "process_rss_before_bytes": 413868032,
  "process_peak_rss_bytes": 670052352,
  "incremental_peak_rss_bytes": 256184320,
  "parameter_count": 65203985,
  "model_directory_bytes": 261501138,
  "environment": {
    "python": "3.12.13",
    "platform": "Windows-11-10.0.26200-SP0",
    "processor": "Intel64 Family 6 Model 186 Stepping 2, GenuineIntel",
    "physical_cpu_cores": 8,
    "logical_cpu_cores": 12,
    "total_ram_bytes": 16890322944,
    "torch": "2.12.0+cpu",
    "transformers": "5.7.0",
    "huggingface_hub": "1.25.1",
    "numpy": "2.5.1",
    "scipy": "1.18.0",
    "scikit_learn": "1.9.0",
    "psutil": "7.2.2",
    "cuda_available": false
  }
}