{
  "label": "GLM-5.3-Flash UD-IQ2_XXS (101.8 GB) reasoning HIGH",
  "tests": {
    "01-blockfall": {
      "error": "DNF: exceeded the 60 minute time limit (partial output kept: 24034 chunks, 82877 thinking chars, 0 answer chars)",
      "partial_thinking_chars": 82877,
      "partial_answer_chars": 0,
      "partial_chunks": 24034
    },
    "03-eruption": {
      "error": "DNF: exceeded the 60 minute time limit (partial output kept: 24055 chunks, 82201 thinking chars, 0 answer chars)",
      "partial_thinking_chars": 82201,
      "partial_answer_chars": 0,
      "partial_chunks": 24055
    },
    "04-the-ledger": {
      "error": "DNF: exceeded the 60 minute time limit (partial output kept: 23828 chunks, 57722 thinking chars, 28979 answer chars)",
      "partial_thinking_chars": 57722,
      "partial_answer_chars": 28979,
      "partial_chunks": 23828
    },
    "05-blind-artist": {
      "error": "DNF: exceeded the 60 minute time limit (partial output kept: 24276 chunks, 84936 thinking chars, 0 answer chars)",
      "partial_thinking_chars": 84936,
      "partial_answer_chars": 0,
      "partial_chunks": 24276
    }
  },
  "vision": {},
  "size_gb_on_disk": 101.8,
  "load_seconds": 30.0,
  "warmup_seconds": 2.4,
  "mem_baseline_gb": 4.54,
  "mem_peak_gb": 102.48,
  "mem_model_gb_est": 97.94,
  "backend": "llama-server direct: ~/repo/tools/llama.cpp-glm5next/build/bin/llama-server",
  "server_timings": [
    {
      "task": 0,
      "prompt_ms": 2120.29,
      "prompt_tokens": 15,
      "prefill_tok_s": 7.07,
      "eval_ms": 215.49,
      "eval_tokens": 3,
      "gen_tok_s": 9.28
    },
    {
      "task": 6
    },
    {
      "task": 24084
    },
    {
      "task": 48198
    },
    {
      "task": 72104
    }
  ]
}