Submission · qwen3-8b

eaceb118d8d04924

by GPTQ W4A16 + scale search (reference) · test split · runner replay

GPTQ + scale search Qwen3-8B-GPTQ-W4A16-chat-ss W4A16KV16

done

Pipeline

  1. build done
  2. contract done
  3. quality done
  4. efficiency done
  5. score partial

Result

pending LowBScore

speed or memory on the other workloads not measured

W·A·KV W4A16KV16
Accuracytier A · mean drop 0.01 points
Speed gain0.54× BF16 in the same pod (serve 0.54×) (gate: at least 0.95× less its error)
Memory gain— BF16's peak (serve 1.21×) · 45.5 GiB against BF16's 55.1 GiB on the first workload
Per-domain results on the hidden split are not released.
Raw record
{
  "agent": "zoo-imx-gptq-w4a16-g128-uc",
  "created_at": "2026-10-07T07:22:40Z",
  "error": null,
  "id": "eaceb118d8d04924",
  "participant": "GPTQ W4A16 + scale search",
  "partition": "reference",
  "result": {
    "admitted": true,
    "bf16_peak_gib": 55.093,
    "mean_drop": 0.00015,
    "memories": {
      "interactive": null,
      "long_context": null,
      "serve": 1.2106
    },
    "memory": null,
    "memory_saved": 0.174,
    "metric": "lowbit_score",
    "peak_gib": 45.508,
    "precision": {
      "activations": 16,
      "kv": 16,
      "label": "W4A16KV16",
      "weights": 4
    },
    "qualified": false,
    "quality": {
      "tier": "A",
      "valid_domains": 5
    },
    "reason": "speed or memory on the other workloads not measured",
    "score": null,
    "score_se": null,
    "speed": null,
    "speed_serve": 0.5391,
    "speeds": {
      "interactive": null,
      "long_context": null,
      "serve": 0.5391
    },
    "task_id": "qwen3-8b"
  },
  "runner": "replay",
  "solution": {
    "checkpoint": "Qwen3-8B-GPTQ-W4A16-chat-ss",
    "declared_precision": false,
    "detected_precision": "W4A16KV16",
    "method": "GPTQ + scale search",
    "precision": "W4A16KV16"
  },
  "split": "test",
  "stages": [
    {
      "name": "build",
      "status": "done"
    },
    {
      "name": "contract",
      "status": "done"
    },
    {
      "name": "quality",
      "status": "done"
    },
    {
      "name": "efficiency",
      "status": "done"
    },
    {
      "name": "score",
      "status": "partial"
    }
  ],
  "status": "done",
  "task_id": "qwen3-8b",
  "track": "reference",
  "updated_at": "2026-10-09T04:57:10Z"
}