{
  "label": "GLM-5.3-Flash UD-IQ1_S (93.1 GB) reasoning MAX, 3 h cap, 262k ctx",
  "tests": {
    "01-blockfall": {
      "error": "DNF: exceeded the 180 minute time limit (partial output kept: 48406 chunks, 233711 thinking chars, 0 answer chars)",
      "partial_thinking_chars": 233711,
      "partial_answer_chars": 0,
      "partial_chunks": 48406
    },
    "03-eruption": {
      "error": "DNF: exceeded the 180 minute time limit (partial output kept: 48375 chunks, 235760 thinking chars, 0 answer chars)",
      "partial_thinking_chars": 235760,
      "partial_answer_chars": 0,
      "partial_chunks": 48375
    },
    "04-the-ledger": {
      "error": "DNF: exceeded the 180 minute time limit (partial output kept: 47831 chunks, 230161 thinking chars, 0 answer chars)",
      "partial_thinking_chars": 230161,
      "partial_answer_chars": 0,
      "partial_chunks": 47831
    },
    "05-blind-artist": {
      "error": "DNF: exceeded the 180 minute time limit (partial output kept: 48358 chunks, 224279 thinking chars, 0 answer chars)",
      "partial_thinking_chars": 224279,
      "partial_answer_chars": 0,
      "partial_chunks": 48358
    }
  },
  "vision": {},
  "size_gb_on_disk": 93.1,
  "load_seconds": 25.0,
  "warmup_seconds": 2.8,
  "mem_baseline_gb": 6.71,
  "mem_peak_gb": 104.85,
  "mem_model_gb_est": 98.14,
  "backend": "llama-server direct: ~/repo/tools/llama.cpp-glm5next/build/bin/llama-server",
  "server_timings": [
    {
      "task": 0,
      "prompt_ms": 2100.89,
      "prompt_tokens": 15,
      "prefill_tok_s": 7.14,
      "eval_ms": 724.76,
      "eval_tokens": 8,
      "gen_tok_s": 9.66
    },
    {
      "task": 11
    },
    {
      "task": 48455
    },
    {
      "task": 96868
    },
    {
      "task": 144752
    }
  ]
}