{
  "model_id": "m2",
  "model_tag": "gemma4:12b",
  "box_class": "cpu-server",
  "started_utc": "2026-08-25T15:08:20Z",
  "nproc": 64,
  "num_predict": 256,
  "prompt_sha256": "90eedd0c53f9554ae3837674504fcb7090013d9a432a653352183c0f25a7ce5c",
  "nvidia_before": "0, 38764 MiB, 0 %",
  "ollama_version": "0.32.15",
  "runs": [
    {
      "ttft_ms": 510.38,
      "first_token_was_thinking": false,
      "wall_s": 24.0252,
      "eval_count": 256,
      "eval_duration_ns": 23509383000,
      "decode_tok_s": 10.889,
      "prompt_eval_count": 536,
      "prompt_eval_duration_ns": 499890000,
      "prefill_tok_s": 1072.236,
      "load_duration_ns": 1449244,
      "total_duration_ns": 24019147923,
      "streamed_chunks": 256,
      "response_chars": 1021,
      "thinking_chars": 0,
      "done_reason": "length",
      "run": 1
    },
    {
      "ttft_ms": 504.2,
      "first_token_was_thinking": false,
      "wall_s": 23.8896,
      "eval_count": 256,
      "eval_duration_ns": 23379996000,
      "decode_tok_s": 10.95,
      "prompt_eval_count": 536,
      "prompt_eval_duration_ns": 495305000,
      "prefill_tok_s": 1082.161,
      "load_duration_ns": 1180345,
      "total_duration_ns": 23883709117,
      "streamed_chunks": 256,
      "response_chars": 1021,
      "thinking_chars": 0,
      "done_reason": "length",
      "run": 2
    },
    {
      "ttft_ms": 503.08,
      "first_token_was_thinking": false,
      "wall_s": 23.813,
      "eval_count": 256,
      "eval_duration_ns": 23304247000,
      "decode_tok_s": 10.985,
      "prompt_eval_count": 536,
      "prompt_eval_duration_ns": 494552000,
      "prefill_tok_s": 1083.809,
      "load_duration_ns": 1100435,
      "total_duration_ns": 23807008469,
      "streamed_chunks": 256,
      "response_chars": 1021,
      "thinking_chars": 0,
      "done_reason": "length",
      "run": 3
    },
    {
      "ttft_ms": 505.54,
      "first_token_was_thinking": false,
      "wall_s": 23.7674,
      "eval_count": 256,
      "eval_duration_ns": 23255737000,
      "decode_tok_s": 11.008,
      "prompt_eval_count": 536,
      "prompt_eval_duration_ns": 495629000,
      "prefill_tok_s": 1081.454,
      "load_duration_ns": 1360614,
      "total_duration_ns": 23760885375,
      "streamed_chunks": 256,
      "response_chars": 1021,
      "thinking_chars": 0,
      "done_reason": "length",
      "run": 4
    },
    {
      "ttft_ms": 503.24,
      "first_token_was_thinking": false,
      "wall_s": 23.7288,
      "eval_count": 256,
      "eval_duration_ns": 23219996000,
      "decode_tok_s": 11.025,
      "prompt_eval_count": 536,
      "prompt_eval_duration_ns": 494109000,
      "prefill_tok_s": 1084.781,
      "load_duration_ns": 1395854,
      "total_duration_ns": 23722890535,
      "streamed_chunks": 256,
      "response_chars": 1021,
      "thinking_chars": 0,
      "done_reason": "length",
      "run": 5
    }
  ],
  "errors": [],
  "quantization": "Q4_K_M",
  "param_size": "11.9B",
  "family": "gemma4",
  "context_length": 262144,
  "think_mode": false,
  "think_note": "think:false accepted",
  "warmup_discarded": {
    "ttft_ms": 494.88,
    "first_token_was_thinking": false,
    "wall_s": 24.2389,
    "eval_count": 256,
    "eval_duration_ns": 23738433000,
    "decode_tok_s": 10.784,
    "prompt_eval_count": 536,
    "prompt_eval_duration_ns": 484498000,
    "prefill_tok_s": 1106.3,
    "load_duration_ns": 1329574,
    "total_duration_ns": 24232948558,
    "streamed_chunks": 256,
    "response_chars": 1021,
    "thinking_chars": 0,
    "done_reason": "length"
  },
  "ps_after_load": [
    {
      "name": "gemma4:12b",
      "model": "gemma4:12b",
      "size": 8897649703,
      "digest": "4eb23ef187e2c5462566d6a1d3bbbc2f1346d0b4327cbb66d58fffbcc9b2b05c",
      "details": {
        "parent_model": "",
        "format": "gguf",
        "family": "gemma4",
        "families": [
          "gemma4"
        ],
        "parameter_size": "11.9B",
        "quantization_level": "Q4_K_M"
      },
      "expires_at": "2026-08-25T15:19:03.268906869Z",
      "size_vram": 0,
      "context_length": 4096
    }
  ],
  "size_vram": 0,
  "size_total": 8897649703,
  "cpu_only_receipt": true,
  "peak_runner_rss_kb": 9522624,
  "peak_runner_rss_gib": 9.08,
  "peak_runner_threads": 103,
  "rss_samples": 218,
  "loadavg_during_runs": {
    "min": 24.82,
    "median": 29.49,
    "max": 31.13,
    "mean": 29.04,
    "samples": 218,
    "nproc": 64
  },
  "contention_suspected": false,
  "contention_note": "loadavg consistent with this bench owning the machine",
  "foreign_gpu_procs_during": "0, 38764 MiB, 0 %",
  "derived": {
    "decode_tok_s": {
      "min": 10.889,
      "median": 10.985,
      "max": 11.025,
      "mean": 10.971,
      "n": 5
    },
    "ttft_ms": {
      "min": 503.08,
      "median": 504.2,
      "max": 510.38,
      "mean": 505.288,
      "n": 5
    },
    "wall_s": {
      "min": 23.7288,
      "median": 23.813,
      "max": 24.0252,
      "mean": 23.845,
      "n": 5
    },
    "prefill_tok_s": {
      "min": 1072.236,
      "median": 1082.161,
      "max": 1084.781,
      "mean": 1080.888,
      "n": 5
    },
    "eval_count": {
      "min": 256,
      "median": 256,
      "max": 256,
      "mean": 256.0,
      "n": 5
    }
  },
  "cold_load": [
    {
      "ttft_ms": 18214.53,
      "first_token_was_thinking": false,
      "wall_s": 18.2184,
      "eval_count": 1,
      "eval_duration_ns": 1000,
      "decode_tok_s": 1000000.0,
      "prompt_eval_count": 536,
      "prompt_eval_duration_ns": 6895143000,
      "prefill_tok_s": 77.736,
      "load_duration_ns": 11304891696,
      "total_duration_ns": 18213572943,
      "streamed_chunks": 1,
      "response_chars": 2,
      "thinking_chars": 0,
      "done_reason": "length",
      "wall_to_first_token_s": 18.2185,
      "ps_confirmed_empty_before": true,
      "load_duration_s": 11.3049
    },
    {
      "ttft_ms": 18202.2,
      "first_token_was_thinking": false,
      "wall_s": 18.2066,
      "eval_count": 1,
      "eval_duration_ns": 1000,
      "decode_tok_s": 1000000.0,
      "prompt_eval_count": 536,
      "prompt_eval_duration_ns": 6851832000,
      "prefill_tok_s": 78.227,
      "load_duration_ns": 11338173148,
      "total_duration_ns": 18201742126,
      "streamed_chunks": 1,
      "response_chars": 2,
      "thinking_chars": 0,
      "done_reason": "length",
      "wall_to_first_token_s": 18.2068,
      "ps_confirmed_empty_before": true,
      "load_duration_s": 11.3382
    }
  ],
  "cold_load_note": "PAGE-CACHE-WARM: no sudo to drop caches, so this is warm-disk reload, not a true cold read from NVMe.",
  "nvidia_after": "0, 38764 MiB, 0 %",
  "ps_final": [
    {
      "name": "gemma4:12b",
      "model": "gemma4:12b",
      "size": 8897649703,
      "digest": "4eb23ef187e2c5462566d6a1d3bbbc2f1346d0b4327cbb66d58fffbcc9b2b05c",
      "details": {
        "parent_model": "",
        "format": "gguf",
        "family": "gemma4",
        "families": [
          "gemma4"
        ],
        "parameter_size": "11.9B",
        "quantization_level": "Q4_K_M"
      },
      "expires_at": "2026-08-25T15:21:41.184029045Z",
      "size_vram": 0,
      "context_length": 4096
    }
  ],
  "finished_utc": "2026-08-25T15:11:42Z"
}