{
  "label": "GLM-5.3-Flash UD-IQ1_S (93.1 GB) reasoning LOW",
  "tests": {
    "01-blockfall": {
      "wall_seconds": 321.8,
      "prompt_tokens": 201,
      "completion_tokens": 3037,
      "tok_s_overall": 9.44,
      "thinking_chars": 0,
      "answer_chars": 9141,
      "thinking_share": 0.0,
      "html_bytes": 9137,
      "gen_tok_s": 9.586147963147374,
      "prefill_tok_s": 45.116545200551265,
      "ttft_est_seconds": 4.26,
      "generate_seconds": 316.7,
      "thinking_seconds_est": 0.0,
      "timings_raw": {
        "cache_n": 9,
        "prompt_n": 192,
        "prompt_ms": 4255.645,
        "prompt_per_token_ms": 22.164817708333334,
        "prompt_per_second": 45.116545200551265,
        "predicted_n": 3037,
        "predicted_ms": 316706.983,
        "predicted_per_token_ms": 104.31718807641634,
        "predicted_per_second": 9.586147963147374
      }
    },
    "03-eruption": {
      "wall_seconds": 483.8,
      "prompt_tokens": 228,
      "completion_tokens": 4426,
      "tok_s_overall": 9.15,
      "thinking_chars": 118,
      "answer_chars": 13162,
      "thinking_share": 0.009,
      "html_bytes": 10993,
      "gen_tok_s": 9.245724845260362,
      "prefill_tok_s": 49.93943418393036,
      "ttft_est_seconds": 4.39,
      "generate_seconds": 478.6,
      "thinking_seconds_est": 4.3,
      "timings_raw": {
        "cache_n": 9,
        "prompt_n": 219,
        "prompt_ms": 4385.312,
        "prompt_per_token_ms": 20.024255707762556,
        "prompt_per_second": 49.93943418393036,
        "predicted_n": 4426,
        "predicted_ms": 478599.577,
        "predicted_per_token_ms": 108.15809649717514,
        "predicted_per_second": 9.245724845260362
      }
    },
    "04-the-ledger": {
      "wall_seconds": 320.4,
      "prompt_tokens": 902,
      "completion_tokens": 2894,
      "tok_s_overall": 9.03,
      "thinking_chars": 419,
      "answer_chars": 7468,
      "thinking_share": 0.053,
      "html_bytes": 7463,
      "gen_tok_s": 9.36776224605172,
      "prefill_tok_s": 82.60143650650497,
      "ttft_est_seconds": 10.81,
      "generate_seconds": 308.8,
      "thinking_seconds_est": 16.4,
      "timings_raw": {
        "cache_n": 9,
        "prompt_n": 893,
        "prompt_ms": 10810.95,
        "prompt_per_token_ms": 12.106326987681971,
        "prompt_per_second": 82.60143650650497,
        "predicted_n": 2894,
        "predicted_ms": 308825.088,
        "predicted_per_token_ms": 106.74907984790875,
        "predicted_per_second": 9.36776224605172
      }
    },
    "05-blind-artist": {
      "wall_seconds": 421.0,
      "prompt_tokens": 190,
      "completion_tokens": 3909,
      "tok_s_overall": 9.29,
      "thinking_chars": 1497,
      "answer_chars": 8508,
      "thinking_share": 0.15,
      "html_bytes": 8496,
      "gen_tok_s": 9.391700654105033,
      "prefill_tok_s": 44.69107111904961,
      "ttft_est_seconds": 4.05,
      "generate_seconds": 416.1,
      "thinking_seconds_est": 62.4,
      "timings_raw": {
        "cache_n": 9,
        "prompt_n": 181,
        "prompt_ms": 4050.026,
        "prompt_per_token_ms": 22.375834254143644,
        "prompt_per_second": 44.69107111904961,
        "predicted_n": 3909,
        "predicted_ms": 416112.07,
        "predicted_per_token_ms": 106.47698822927329,
        "predicted_per_second": 9.391700654105033
      }
    }
  },
  "vision": {},
  "size_gb_on_disk": 93.1,
  "load_seconds": 25.0,
  "warmup_seconds": 2.9,
  "mem_baseline_gb": 6.43,
  "mem_peak_gb": 95.61,
  "mem_model_gb_est": 89.18,
  "backend": "llama-server direct: ~/private/jose-yt/tools/llama.cpp-glm5next/build/bin/llama-server",
  "server_timings": [
    {
      "task": 0,
      "prompt_ms": 795.6,
      "prompt_tokens": 15,
      "prefill_tok_s": 18.85,
      "eval_ms": 732.04,
      "eval_tokens": 8,
      "gen_tok_s": 9.56
    },
    {
      "task": 11,
      "prompt_ms": 2939.42,
      "prompt_tokens": 192,
      "prefill_tok_s": 65.32,
      "eval_ms": 318727.0,
      "eval_tokens": 3037,
      "gen_tok_s": 9.53
    },
    {
      "task": 3050,
      "prompt_ms": 3143.75,
      "prompt_tokens": 219,
      "prefill_tok_s": 69.66,
      "eval_ms": 479941.82,
      "eval_tokens": 4426,
      "gen_tok_s": 9.22
    },
    {
      "task": 7478,
      "prompt_ms": 8983.5,
      "prompt_tokens": 893,
      "prefill_tok_s": 99.4,
      "eval_ms": 309712.69,
      "eval_tokens": 2894,
      "gen_tok_s": 9.34
    },
    {
      "task": 10375,
      "prompt_ms": 2836.72,
      "prompt_tokens": 181,
      "prefill_tok_s": 63.81,
      "eval_ms": 416247.18,
      "eval_tokens": 3909,
      "gen_tok_s": 9.39
    },
    {
      "task": 0,
      "prompt_ms": 2151.19,
      "prompt_tokens": 15,
      "prefill_tok_s": 6.97,
      "eval_ms": 728.59,
      "eval_tokens": 8,
      "gen_tok_s": 9.61
    },
    {
      "task": 11,
      "prompt_ms": 4255.65,
      "prompt_tokens": 192,
      "prefill_tok_s": 45.12,
      "eval_ms": 316706.98,
      "eval_tokens": 3037,
      "gen_tok_s": 9.59
    },
    {
      "task": 3050,
      "prompt_ms": 4385.31,
      "prompt_tokens": 219,
      "prefill_tok_s": 49.94,
      "eval_ms": 478599.58,
      "eval_tokens": 4426,
      "gen_tok_s": 9.25
    },
    {
      "task": 7478,
      "prompt_ms": 10810.95,
      "prompt_tokens": 893,
      "prefill_tok_s": 82.6,
      "eval_ms": 308825.09,
      "eval_tokens": 2894,
      "gen_tok_s": 9.37
    },
    {
      "task": 10375,
      "prompt_ms": 4050.03,
      "prompt_tokens": 181,
      "prefill_tok_s": 44.69,
      "eval_ms": 416112.07,
      "eval_tokens": 3909,
      "gen_tok_s": 9.39
    }
  ]
}