{
  "label": "GLM-5.3-Flash UD-IQ1_S (93.1 GB) reasoning MAX",
  "tests": {
    "01-blockfall": {
      "error": "DNF: exceeded the 60 minute time limit (partial output kept: 23888 chunks, 86603 thinking chars, 0 answer chars)",
      "partial_thinking_chars": 86603,
      "partial_answer_chars": 0,
      "partial_chunks": 23888
    },
    "03-eruption": {
      "error": "DNF: exceeded the 60 minute time limit (partial output kept: 24523 chunks, 92648 thinking chars, 0 answer chars)",
      "partial_thinking_chars": 92648,
      "partial_answer_chars": 0,
      "partial_chunks": 24523
    },
    "04-the-ledger": {
      "error": "DNF: exceeded the 60 minute time limit (partial output kept: 24182 chunks, 88267 thinking chars, 0 answer chars)",
      "partial_thinking_chars": 88267,
      "partial_answer_chars": 0,
      "partial_chunks": 24182
    },
    "05-blind-artist": {
      "error": "DNF: exceeded the 60 minute time limit (partial output kept: 24618 chunks, 81839 thinking chars, 0 answer chars)",
      "partial_thinking_chars": 81839,
      "partial_answer_chars": 0,
      "partial_chunks": 24618
    }
  },
  "vision": {},
  "size_gb_on_disk": 93.1,
  "load_seconds": 25.0,
  "warmup_seconds": 2.9,
  "mem_baseline_gb": 6.02,
  "mem_peak_gb": 101.05,
  "mem_model_gb_est": 95.03,
  "backend": "llama-server direct: ~/repo/tools/llama.cpp-glm5next/build/bin/llama-server",
  "server_timings": [
    {
      "task": 0,
      "prompt_ms": 2132.3,
      "prompt_tokens": 15,
      "prefill_tok_s": 7.03,
      "eval_ms": 733.88,
      "eval_tokens": 8,
      "gen_tok_s": 9.54
    },
    {
      "task": 11
    },
    {
      "task": 23937
    },
    {
      "task": 48498
    },
    {
      "task": 72733
    }
  ]
}