{
  "model": "Qwen/Qwen3.5-4B",
  "packages": {
    "torch": "2.11.0",
    "vllm": "0.26.0",
    "transformers": "5.18.0",
    "tokenizers": "0.23.2",
    "huggingface_hub": "1.33.0"
  },
  "args": {
    "input": "data/cliff_realistic_dense_v1/requests.jsonl",
    "output": "results/cliff_realistic_dense_v1_4b_native.jsonl",
    "model": "Qwen/Qwen3.5-4B",
    "revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
    "engine": "vllm",
    "batch_size": 12,
    "max_tokens": 32768,
    "max_model_len": 131072,
    "gpu_memory_utilization": 0.85,
    "thinking": "on",
    "limit": null,
    "cache_dir": "cache/huggingface",
    "enforce_eager": false
  },
  "engine_environment": {
    "VLLM_USE_FLASHINFER_SAMPLER": "0"
  },
  "started_unix": 1790848851.5893328,
  "pending_samples": 504,
  "thinking_enabled": true,
  "completed_samples_before_run": 0,
  "dtype": "bfloat16",
  "hardware": {
    "cuda": "13.0",
    "device": "NVIDIA RTX PRO 6000 Blackwell Server Edition",
    "total_memory_bytes": 101975851008,
    "nvidia_smi": "Thu Oct  1 10:00:51 2026       \n+-----------------------------------------------------------------------------------------+\n| NVIDIA-SMI 595.71.05              Driver Version: 595.71.05      CUDA Version: 13.2     |\n+-----------------------------------------+------------------------+----------------------+\n| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |\n| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |\n|                                         |                        |               MIG M. |\n|=========================================+========================+======================|\n|   0  NVIDIA RTX PRO 6000 Blac...    Off |   00000000:20:00.0 Off |                    0 |\n| N/A   26C    P0             56W /  600W |       3MiB /  97887MiB |      0%      Default |\n|                                         |                        |             Disabled |\n+-----------------------------------------+------------------------+----------------------+\n\n+-----------------------------------------------------------------------------------------+\n| Processes:                                                                              |\n|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |\n|        ID   ID                                                               Usage      |\n|=========================================================================================|\n|  No running processes found                                                             |\n+-----------------------------------------------------------------------------------------+\n"
  },
  "sampling": {
    "temperature": 1.0,
    "top_p": 0.95,
    "top_k": 20,
    "min_p": 0.0,
    "presence_penalty": 1.5,
    "repetition_penalty": 1.0
  },
  "latency_definition": "vLLM finished_time minus arrival_time when exposed; otherwise batch wall time",
  "load_mode": "language_model_only",
  "sampling_profile": "official_model_mode",
  "model_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
  "revision_source": "loaded config cache snapshot identifier",
  "decoding": {
    "skip_special_tokens": false,
    "preserve_think_delimiters": true
  },
  "model_load_seconds": 48.20975907100001,
  "generated_samples": 504,
  "generated_tokens": 6528329,
  "generation_seconds": 13084.019918500999,
  "last_batch_seconds": 334.07184085600056,
  "elapsed_seconds": 13132.317841529846,
  "tokens_per_second": 498.95437645802156,
  "seconds_per_sample": 25.960356981152774,
  "completed_unix": 1790861983.9080038
}