{
  "label": "GLM-5.3-Flash UD-IQ2_XXS (101.8 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": 440.9,
      "prompt_tokens": 902,
      "completion_tokens": 3863,
      "tok_s_overall": 8.76,
      "thinking_chars": 1268,
      "answer_chars": 8926,
      "thinking_share": 0.124,
      "html_bytes": 8924,
      "gen_tok_s": 9.024065980605702,
      "prefill_tok_s": 81.74431568061938,
      "ttft_est_seconds": 10.92,
      "generate_seconds": 428.0,
      "thinking_seconds_est": 53.1,
      "timings_raw": {
        "cache_n": 9,
        "prompt_n": 893,
        "prompt_ms": 10924.307,
        "prompt_per_token_ms": 12.233266517357224,
        "prompt_per_second": 81.74431568061938,
        "predicted_n": 3863,
        "predicted_ms": 427966.729,
        "predicted_per_token_ms": 110.81479259451062,
        "predicted_per_second": 9.024065980605702
      }
    },
    "05-blind-artist": {
      "wall_seconds": 444.2,
      "prompt_tokens": 190,
      "completion_tokens": 4061,
      "tok_s_overall": 9.14,
      "thinking_chars": 124,
      "answer_chars": 10168,
      "thinking_share": 0.012,
      "html_bytes": 9632,
      "gen_tok_s": 9.24692865029996,
      "prefill_tok_s": 43.74130762343326,
      "ttft_est_seconds": 4.14,
      "generate_seconds": 439.1,
      "thinking_seconds_est": 5.3,
      "timings_raw": {
        "cache_n": 9,
        "prompt_n": 181,
        "prompt_ms": 4137.965,
        "prompt_per_token_ms": 22.861685082872928,
        "prompt_per_second": 43.74130762343326,
        "predicted_n": 4061,
        "predicted_ms": 439064.705,
        "predicted_per_token_ms": 108.14401600985222,
        "predicted_per_second": 9.24692865029996
      }
    }
  },
  "vision": {},
  "size_gb_on_disk": 101.8,
  "load_seconds": 30.0,
  "warmup_seconds": 2.5,
  "mem_baseline_gb": 6.3,
  "mem_peak_gb": 103.42,
  "mem_model_gb_est": 97.12,
  "backend": "llama-server direct: ~/repo/tools/llama.cpp-glm5next/build/bin/llama-server",
  "server_timings": [
    {
      "task": 0,
      "prompt_ms": 2219.17,
      "prompt_tokens": 15,
      "prefill_tok_s": 6.76,
      "eval_ms": 216.09,
      "eval_tokens": 3,
      "gen_tok_s": 9.26
    },
    {
      "task": 6
    },
    {
      "task": 24128
    },
    {
      "task": 48263,
      "prompt_ms": 10924.31,
      "prompt_tokens": 893,
      "prefill_tok_s": 81.74,
      "eval_ms": 427966.73,
      "eval_tokens": 3863,
      "gen_tok_s": 9.02
    },
    {
      "task": 52135,
      "prompt_ms": 4137.97,
      "prompt_tokens": 181,
      "prefill_tok_s": 43.74,
      "eval_ms": 439064.71,
      "eval_tokens": 4061,
      "gen_tok_s": 9.25
    }
  ]
}