`battle/80b.json`

150 lines
{
  "model": "mlx-community/Qwen3-Next-80B-A3B-Instruct-4bit",
  "when": "2026-08-18T16:59:04.261Z",
  "score": "12/12",
  "passed": 12,
  "total": 12,
  "tok_per_s": 21.4,
  "tok_s_ms": 7526,
  "longctx": {
    "placed": true,
    "pass": true,
    "got": "MAUVE-7719",
    "ms": 27782,
    "approx_ctx_tokens": 33990
  },
  "byCat": {
    "json": {
      "pass": 3,
      "tot": 3
    },
    "classify": {
      "pass": 3,
      "tot": 3
    },
    "constraint": {
      "pass": 2,
      "tot": 2
    },
    "reason": {
      "pass": 2,
      "tot": 2
    },
    "code": {
      "pass": 2,
      "tot": 2
    }
  },
  "results": [
    {
      "id": "json_person",
      "cat": "json",
      "pass": true,
      "got": {
        "name": "Maria Gonzalez",
        "age": 34,
        "city": "Chicago"
      },
      "ms": 2179,
      "out": "{\"name\": \"Maria Gonzalez\", \"age\": 34, \"city\": \"Chicago\"}"
    },
    {
      "id": "json_order",
      "cat": "json",
      "pass": true,
      "got": {
        "sku": "CORK-500210",
        "qty": 3,
        "price_usd": 59.5
      },
      "ms": 3071,
      "out": "{\"sku\": \"CORK-500210\", \"qty\": 3, \"price_usd\": 59.5}"
    },
    {
      "id": "json_list",
      "cat": "json",
      "pass": true,
      "got": {
        "items": [
          "silk wallpaper",
          "cork tiles",
          "grasscloth panels"
        ]
      },
      "ms": 1587,
      "out": "{\"items\": [\"silk wallpaper\", \"cork tiles\", \"grasscloth panels\"]}"
    },
    {
      "id": "cls_sentiment",
      "cat": "classify",
      "pass": true,
      "got": "negative",
      "ms": 1074,
      "out": "negative"
    },
    {
      "id": "cls_ticket",
      "cat": "classify",
      "pass": true,
      "got": "billing",
      "ms": 1537,
      "out": "billing"
    },
    {
      "id": "cls_spam",
      "cat": "classify",
      "pass": true,
      "got": "yes",
      "ms": 756,
      "out": "yes"
    },
    {
      "id": "cons_oneword",
      "cat": "constraint",
      "pass": true,
      "got": "Paris",
      "ms": 787,
      "out": "Paris"
    },
    {
      "id": "cons_numonly",
      "cat": "constraint",
      "pass": true,
      "got": "391",
      "ms": 1300,
      "out": "391"
    },
    {
      "id": "reason_word",
      "cat": "reason",
      "pass": true,
      "got": "wall:\n\n$ 135 - 108 = 27 $ sq ft.\n\n### ✅ Final Answer: **27**",
      "ms": 6545,
      "out": "We are given:\n\n- Each roll covers **27 sq ft**.\n- There are **5 rolls**, so total coverage available is:  \n  $ 5 \\times 27 = 135 $ sq ft.\n\n- The wall is **12 ft"
    },
    {
      "id": "reason_logic",
      "cat": "reason",
      "pass": true,
      "got": "Mia",
      "ms": 891,
      "out": "Mia"
    },
    {
      "id": "code_sum",
      "cat": "code",
      "pass": true,
      "got": "function sumTo(n){let s=0;for(let i=1;i<=n;i++)s+=i;return s;}",
      "ms": 1586,
      "out": "```js\nfunction sumTo(n){let s=0;for(let i=1;i<=n;i++)s+=i;return s;}\n```"
    },
    {
      "id": "code_null",
      "cat": "code",
      "pass": true,
      "got": "function total(o){return (o.price || 0) * (o.qty || 0);}",
      "ms": 2700,
      "out": "```js\nfunction total(o){return (o.price || 0) * (o.qty || 0);}\n```"
    }
  ]
}