{
  "label": "GLM-5.3-Flash UD-Q2_K_XL (108.7 GB) reasoning LOW",
  "tests": {
    "01-blockfall": {
      "wall_seconds": 337.2,
      "prompt_tokens": 201,
      "completion_tokens": 3074,
      "tok_s_overall": 9.12,
      "thinking_chars": 41,
      "answer_chars": 9179,
      "thinking_share": 0.004,
      "html_bytes": 9167,
      "gen_tok_s": 9.197070927146575,
      "prefill_tok_s": 64.31697549422738,
      "ttft_est_seconds": 2.99,
      "generate_seconds": 334.1,
      "thinking_seconds_est": 1.3,
      "timings_raw": {
        "cache_n": 9,
        "prompt_n": 192,
        "prompt_ms": 2985.215,
        "prompt_per_token_ms": 15.547994791666667,
        "prompt_per_second": 64.31697549422738,
        "predicted_n": 3074,
        "predicted_ms": 334128.118,
        "predicted_per_token_ms": 108.73026944354052,
        "predicted_per_second": 9.197070927146575
      }
    },
    "03-eruption": {
      "error": "DNF: exceeded the 60 minute time limit"
    },
    "04-the-ledger": {
      "wall_seconds": 611.4,
      "prompt_tokens": 902,
      "completion_tokens": 5193,
      "tok_s_overall": 8.49,
      "thinking_chars": 1365,
      "answer_chars": 14479,
      "thinking_share": 0.086,
      "html_bytes": 14492,
      "gen_tok_s": 8.635361289886403,
      "prefill_tok_s": 95.64520833036562,
      "ttft_est_seconds": 9.34,
      "generate_seconds": 601.2,
      "thinking_seconds_est": 51.7,
      "timings_raw": {
        "cache_n": 9,
        "prompt_n": 893,
        "prompt_ms": 9336.589,
        "prompt_per_token_ms": 10.455306830907055,
        "prompt_per_second": 95.64520833036562,
        "predicted_n": 5193,
        "predicted_ms": 601248.729,
        "predicted_per_token_ms": 115.80291390600925,
        "predicted_per_second": 8.635361289886403
      }
    },
    "05-blind-artist": {
      "wall_seconds": 1125.6,
      "prompt_tokens": 190,
      "completion_tokens": 9185,
      "tok_s_overall": 8.16,
      "thinking_chars": 15989,
      "answer_chars": 9303,
      "thinking_share": 0.632,
      "html_bytes": 7857,
      "gen_tok_s": 8.182999919953794,
      "prefill_tok_s": 59.32461709401149,
      "ttft_est_seconds": 3.05,
      "generate_seconds": 1122.3,
      "thinking_seconds_est": 709.3,
      "timings_raw": {
        "cache_n": 9,
        "prompt_n": 181,
        "prompt_ms": 3051.01,
        "prompt_per_token_ms": 16.856408839779007,
        "prompt_per_second": 59.32461709401149,
        "predicted_n": 9185,
        "predicted_ms": 1122326.786,
        "predicted_per_token_ms": 122.20457164634148,
        "predicted_per_second": 8.182999919953794
      }
    }
  },
  "vision": {},
  "size_gb_on_disk": 108.7,
  "load_seconds": 30.0,
  "warmup_seconds": 1.1,
  "mem_baseline_gb": 7.25,
  "mem_peak_gb": 111.27,
  "mem_model_gb_est": 104.02,
  "backend": "llama-server direct: ~/repo/tools/llama.cpp-glm5next/build/bin/llama-server",
  "server_timings": [
    {
      "task": 0,
      "prompt_ms": 794.69,
      "prompt_tokens": 15,
      "prefill_tok_s": 18.88,
      "eval_ms": 210.17,
      "eval_tokens": 3,
      "gen_tok_s": 9.52
    },
    {
      "task": 6,
      "prompt_ms": 3013.59,
      "prompt_tokens": 192,
      "prefill_tok_s": 63.71,
      "eval_ms": 335316.85,
      "eval_tokens": 3074,
      "gen_tok_s": 9.16
    },
    {
      "task": 3082
    },
    {
      "task": 58969,
      "prompt_ms": 9524.0,
      "prompt_tokens": 893,
      "prefill_tok_s": 93.76,
      "eval_ms": 602809.74,
      "eval_tokens": 5193,
      "gen_tok_s": 8.61
    },
    {
      "task": 64167,
      "prompt_ms": 3118.42,
      "prompt_tokens": 181,
      "prefill_tok_s": 58.04,
      "eval_ms": 1113293.44,
      "eval_tokens": 9185,
      "gen_tok_s": 8.25
    },
    {
      "task": 0,
      "prompt_ms": 790.07,
      "prompt_tokens": 15,
      "prefill_tok_s": 18.99,
      "eval_ms": 220.66,
      "eval_tokens": 3,
      "gen_tok_s": 9.06
    },
    {
      "task": 6,
      "prompt_ms": 2985.22,
      "prompt_tokens": 192,
      "prefill_tok_s": 64.32,
      "eval_ms": 334128.12,
      "eval_tokens": 3074,
      "gen_tok_s": 9.2
    },
    {
      "task": 3082
    },
    {
      "task": 58957,
      "prompt_ms": 9336.59,
      "prompt_tokens": 893,
      "prefill_tok_s": 95.65,
      "eval_ms": 601248.73,
      "eval_tokens": 5193,
      "gen_tok_s": 8.64
    },
    {
      "task": 64154,
      "prompt_ms": 3051.01,
      "prompt_tokens": 181,
      "prefill_tok_s": 59.32,
      "eval_ms": 1122326.79,
      "eval_tokens": 9185,
      "gen_tok_s": 8.18
    },
    {
      "task": 0,
      "prompt_ms": 829.81,
      "prompt_tokens": 15,
      "prefill_tok_s": 18.08,
      "eval_ms": 221.99,
      "eval_tokens": 3,
      "gen_tok_s": 9.01
    },
    {
      "task": 6
    }
  ]
}