Submission · qwen3-8b

`8592ee4aa17b4140`

by GPTQ W4A16 (reference) · test split · runner replay

GPTQ Qwen3-8B-GPTQ-W4A16 W4A16KV16

done

Pipeline

build done
contract done
quality done
efficiency done
score done

Result

0.073LowBScore

accuracy gate: the worst domain's 95% upper bound on the drop is 5.77 points, over 5

W·A·KV	W4A16KV16
Accuracy	fail · mean drop 2.36 points
Speed gain	0.66× BF16 in the same pod (serve 0.54×, interactive 1.17×, long context 0.45×) (gate: at least 0.95× less its error)
Memory gain	1.43× BF16's peak (serve 1.21×, interactive 2.29×, long context 1.06×) · 45.5 GiB against BF16's 55.1 GiB on the first workload
Efficiency gain E	0.94
Per-domain results on the hidden split are not released.

Raw record

{
  "agent": "zoo-gptq-w4a16-g128",
  "created_at": "2026-10-07T07:22:39Z",
  "error": null,
  "id": "8592ee4aa17b4140",
  "participant": "GPTQ W4A16",
  "partition": "reference",
  "result": {
    "admitted": false,
    "bf16_peak_gib": 55.097,
    "drift": 2.54,
    "efficiency_gain": 0.9431,
    "length_shift": 0.06116,
    "mean_drop": 0.02364,
    "memories": {
      "interactive": 2.285,
      "long_context": 1.0641,
      "serve": 1.2113
    },
    "memory": 1.4334,
    "memory_saved": 0.1744,
    "metric": "lowbit_score",
    "peak_gib": 45.486,
    "precision": {
      "activations": 16,
      "kv": 16,
      "label": "W4A16KV16",
      "weights": 4
    },
    "qualified": false,
    "quality": {
      "tier": "fail",
      "valid_domains": 5
    },
    "reason": "accuracy gate: the worst domain\u0027s 95% upper bound on the drop is 5.77 points, over 5",
    "score": 0.0732,
    "score_se": 0.0027,
    "speed": 0.6585,
    "speed_serve": 0.5443,
    "speeds": {
      "interactive": 1.1749,
      "long_context": 0.4465,
      "serve": 0.5443
    },
    "task_id": "qwen3-8b"
  },
  "runner": "replay",
  "solution": {
    "checkpoint": "Qwen3-8B-GPTQ-W4A16",
    "declared_precision": false,
    "detected_precision": "W4A16KV16",
    "method": "GPTQ",
    "precision": "W4A16KV16"
  },
  "split": "test",
  "stages": [
    {
      "name": "build",
      "status": "done"
    },
    {
      "name": "contract",
      "status": "done"
    },
    {
      "name": "quality",
      "status": "done"
    },
    {
      "name": "efficiency",
      "status": "done"
    },
    {
      "name": "score",
      "status": "done"
    }
  ],
  "status": "done",
  "task_id": "qwen3-8b",
  "track": "reference",
  "updated_at": "2026-10-09T04:57:10Z"
}