{
  "model_id": "m4",
  "model_tag": "mistral-small3.2:24b",
  "box_class": "cpu-server",
  "started_utc": "2026-08-25T15:15:10Z",
  "nproc": 64,
  "num_predict": 256,
  "prompt_sha256": "90eedd0c53f9554ae3837674504fcb7090013d9a432a653352183c0f25a7ce5c",
  "nvidia_before": "0, 38764 MiB, 0 %",
  "ollama_version": "0.32.15",
  "runs": [
    {
      "ttft_ms": 154.47,
      "first_token_was_thinking": false,
      "wall_s": 36.5805,
      "eval_count": 256,
      "eval_duration_ns": 36422856000,
      "decode_tok_s": 7.029,
      "prompt_eval_count": 1023,
      "prompt_eval_duration_ns": 145547000,
      "prefill_tok_s": 7028.657,
      "load_duration_ns": 1229145,
      "total_duration_ns": 36576980658,
      "streamed_chunks": 256,
      "response_chars": 1036,
      "thinking_chars": 0,
      "done_reason": "length",
      "run": 1
    },
    {
      "ttft_ms": 151.3,
      "first_token_was_thinking": false,
      "wall_s": 36.2575,
      "eval_count": 256,
      "eval_duration_ns": 36102749000,
      "decode_tok_s": 7.091,
      "prompt_eval_count": 1023,
      "prompt_eval_duration_ns": 143182000,
      "prefill_tok_s": 7144.753,
      "load_duration_ns": 1397644,
      "total_duration_ns": 36253702500,
      "streamed_chunks": 256,
      "response_chars": 1036,
      "thinking_chars": 0,
      "done_reason": "length",
      "run": 2
    },
    {
      "ttft_ms": 150.07,
      "first_token_was_thinking": false,
      "wall_s": 36.2829,
      "eval_count": 256,
      "eval_duration_ns": 36129896000,
      "decode_tok_s": 7.086,
      "prompt_eval_count": 1023,
      "prompt_eval_duration_ns": 141807000,
      "prefill_tok_s": 7214.03,
      "load_duration_ns": 1221564,
      "total_duration_ns": 36279495865,
      "streamed_chunks": 256,
      "response_chars": 1036,
      "thinking_chars": 0,
      "done_reason": "length",
      "run": 3
    },
    {
      "ttft_ms": 150.07,
      "first_token_was_thinking": false,
      "wall_s": 35.9669,
      "eval_count": 256,
      "eval_duration_ns": 35814139000,
      "decode_tok_s": 7.148,
      "prompt_eval_count": 1023,
      "prompt_eval_duration_ns": 142133000,
      "prefill_tok_s": 7197.484,
      "load_duration_ns": 1301374,
      "total_duration_ns": 35963671344,
      "streamed_chunks": 256,
      "response_chars": 1036,
      "thinking_chars": 0,
      "done_reason": "length",
      "run": 4
    },
    {
      "ttft_ms": 149.43,
      "first_token_was_thinking": false,
      "wall_s": 35.9005,
      "eval_count": 256,
      "eval_duration_ns": 35748016000,
      "decode_tok_s": 7.161,
      "prompt_eval_count": 1023,
      "prompt_eval_duration_ns": 141356000,
      "prefill_tok_s": 7237.047,
      "load_duration_ns": 1095745,
      "total_duration_ns": 35897261011,
      "streamed_chunks": 256,
      "response_chars": 1036,
      "thinking_chars": 0,
      "done_reason": "length",
      "run": 5
    }
  ],
  "errors": [],
  "quantization": "Q4_K_M",
  "param_size": "24.0B",
  "family": "mistral3",
  "context_length": 131072,
  "think_mode": false,
  "think_note": "think:false accepted",
  "warmup_discarded": {
    "ttft_ms": 155.37,
    "first_token_was_thinking": false,
    "wall_s": 37.1695,
    "eval_count": 256,
    "eval_duration_ns": 37011716000,
    "decode_tok_s": 6.917,
    "prompt_eval_count": 1023,
    "prompt_eval_duration_ns": 146932000,
    "prefill_tok_s": 6962.404,
    "load_duration_ns": 1350554,
    "total_duration_ns": 37166499486,
    "streamed_chunks": 256,
    "response_chars": 1036,
    "thinking_chars": 0,
    "done_reason": "length"
  },
  "ps_after_load": [
    {
      "name": "mistral-small3.2:24b",
      "model": "mistral-small3.2:24b",
      "size": 15305979984,
      "digest": "5a408ab55df5c1b5cf46533c368813b30bf9e4d8fc39263bf2a3338cfa3b895b",
      "details": {
        "parent_model": "",
        "format": "gguf",
        "family": "mistral3",
        "families": [
          "mistral3"
        ],
        "parameter_size": "24.0B",
        "quantization_level": "Q4_K_M"
      },
      "expires_at": "2026-08-25T15:26:31.533929168Z",
      "size_vram": 0,
      "context_length": 4096
    }
  ],
  "size_vram": 0,
  "size_total": 15305979984,
  "cpu_only_receipt": true,
  "peak_runner_rss_kb": 15838832,
  "peak_runner_rss_gib": 15.11,
  "peak_runner_threads": 103,
  "rss_samples": 330,
  "loadavg_during_runs": {
    "min": 26.05,
    "median": 30.9,
    "max": 31.9,
    "mean": 30.25,
    "samples": 330,
    "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": 7.029,
      "median": 7.091,
      "max": 7.161,
      "mean": 7.103,
      "n": 5
    },
    "ttft_ms": {
      "min": 149.43,
      "median": 150.07,
      "max": 154.47,
      "mean": 151.068,
      "n": 5
    },
    "wall_s": {
      "min": 35.9005,
      "median": 36.2575,
      "max": 36.5805,
      "mean": 36.198,
      "n": 5
    },
    "prefill_tok_s": {
      "min": 7028.657,
      "median": 7197.484,
      "max": 7237.047,
      "mean": 7164.394,
      "n": 5
    },
    "eval_count": {
      "min": 256,
      "median": 256,
      "max": 256,
      "mean": 256.0,
      "n": 5
    }
  },
  "cold_load": [
    {
      "ttft_ms": 43170.25,
      "first_token_was_thinking": false,
      "wall_s": 43.1728,
      "eval_count": 1,
      "eval_duration_ns": 1000,
      "decode_tok_s": 1000000.0,
      "prompt_eval_count": 1023,
      "prompt_eval_duration_ns": 24634502000,
      "prefill_tok_s": 41.527,
      "load_duration_ns": 18522275505,
      "total_duration_ns": 43169225526,
      "streamed_chunks": 1,
      "response_chars": 4,
      "thinking_chars": 0,
      "done_reason": "length",
      "wall_to_first_token_s": 43.173,
      "ps_confirmed_empty_before": true,
      "load_duration_s": 18.5223
    },
    {
      "ttft_ms": 42853.76,
      "first_token_was_thinking": false,
      "wall_s": 42.8558,
      "eval_count": 1,
      "eval_duration_ns": 1000,
      "decode_tok_s": 1000000.0,
      "prompt_eval_count": 1023,
      "prompt_eval_duration_ns": 24608311000,
      "prefill_tok_s": 41.571,
      "load_duration_ns": 18235828135,
      "total_duration_ns": 42852369573,
      "streamed_chunks": 1,
      "response_chars": 4,
      "thinking_chars": 0,
      "done_reason": "length",
      "wall_to_first_token_s": 42.856,
      "ps_confirmed_empty_before": true,
      "load_duration_s": 18.2358
    }
  ],
  "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": "mistral-small3.2:24b",
      "model": "mistral-small3.2:24b",
      "size": 15305979984,
      "digest": "5a408ab55df5c1b5cf46533c368813b30bf9e4d8fc39263bf2a3338cfa3b895b",
      "details": {
        "parent_model": "",
        "format": "gguf",
        "family": "mistral3",
        "families": [
          "mistral3"
        ],
        "parameter_size": "24.0B",
        "quantization_level": "Q4_K_M"
      },
      "expires_at": "2026-08-25T15:31:02.569999498Z",
      "size_vram": 0,
      "context_length": 4096
    }
  ],
  "finished_utc": "2026-08-25T15:21:04Z"
}