{
  "label": "GLM-5.3-Flash UD-IQ1_M (97.6 GB) reasoning LOW",
  "tests": {
    "01-blockfall": {
      "error": "DNF: exceeded the 60 minute time limit"
    },
    "03-eruption": {
      "error": "DNF: exceeded the 60 minute time limit"
    },
    "04-the-ledger": {
      "wall_seconds": 498.0,
      "prompt_tokens": 902,
      "completion_tokens": 4355,
      "tok_s_overall": 8.74,
      "thinking_chars": 2670,
      "answer_chars": 9213,
      "thinking_share": 0.225,
      "html_bytes": 9204,
      "gen_tok_s": 8.925984063194623,
      "prefill_tok_s": 96.9456686292827,
      "ttft_est_seconds": 9.21,
      "generate_seconds": 487.8,
      "thinking_seconds_est": 109.8,
      "timings_raw": {
        "cache_n": 9,
        "prompt_n": 893,
        "prompt_ms": 9211.345,
        "prompt_per_token_ms": 10.315055991041433,
        "prompt_per_second": 96.9456686292827,
        "predicted_n": 4355,
        "predicted_ms": 487789.354,
        "predicted_per_token_ms": 112.03246531924667,
        "predicted_per_second": 8.925984063194623
      }
    },
    "05-blind-artist": {
      "error": "DNF: exceeded the 60 minute time limit"
    }
  },
  "vision": {},
  "size_gb_on_disk": 97.6,
  "load_seconds": 30.0,
  "warmup_seconds": 2.4,
  "mem_baseline_gb": 11.66,
  "mem_peak_gb": 110.85,
  "mem_model_gb_est": 99.19,
  "backend": "llama-server direct: ~/repo/tools/llama.cpp-glm5next/build/bin/llama-server",
  "server_timings": [
    {
      "task": 0,
      "prompt_ms": 770.08,
      "prompt_tokens": 15,
      "prefill_tok_s": 19.48,
      "eval_ms": 217.23,
      "eval_tokens": 3,
      "gen_tok_s": 9.21
    },
    {
      "task": 6
    },
    {
      "task": 56239
    },
    {
      "task": 112460,
      "prompt_ms": 9211.34,
      "prompt_tokens": 893,
      "prefill_tok_s": 96.95,
      "eval_ms": 487789.35,
      "eval_tokens": 4355,
      "gen_tok_s": 8.93
    },
    {
      "task": 116820
    },
    {
      "task": 0,
      "prompt_ms": 2185.91,
      "prompt_tokens": 15,
      "prefill_tok_s": 6.86,
      "eval_ms": 220.44,
      "eval_tokens": 3,
      "gen_tok_s": 9.07
    },
    {
      "task": 6
    },
    {
      "task": 24358
    },
    {
      "task": 48735
    }
  ]
}