Submission · qwen35-27b

`aa936ba3e3dd4141`

by AxionML/Qwen3.5-27B-NVFP4 (reference) · test split · runner kubernetes

NVFP4, ModelOpt AxionML/Qwen3.5-27B-NVFP4 W4A4KV16

done

Pipeline

build done
contract done
quality done
efficiency done
score done

Result

pending LowBScore

speed or memory on the other workloads not measured

W·A·KV	W4A4KV16
Accuracy	tier A · mean drop 0.95 points
Speed gain	1.74× BF16 in the same pod (serve 1.74×) (gate: at least 0.95× less its error)
Memory gain	— BF16's peak (serve 1.77×) · 44.0 GiB against BF16's 77.8 GiB on the first workload
Per-domain results on the hidden split are not released.

1.23 GPU-hours

Raw record

{
  "agent": "hf.AxionML.Qwen3.5-27B-NVFP4",
  "created_at": "2026-10-07T02:00:48Z",
  "error": null,
  "id": "aa936ba3e3dd4141",
  "participant": "AxionML/Qwen3.5-27B-NVFP4",
  "partition": "reference",
  "result": {
    "admitted": true,
    "anchor_efficiency": {
      "serve": 831.4987694904767
    },
    "bf16_peak_gib": 77.756,
    "efficiency": {
      "serve": 1449.977624959576
    },
    "gpu_hours": 1.234,
    "mean_drop": 0.00949,
    "memories": {
      "interactive": null,
      "long_context": null,
      "serve": 1.7663
    },
    "memory": null,
    "memory_saved": 0.4338,
    "metric": "lowbit_score",
    "peak_gib": 44.022,
    "peak_memory": {
      "serve": {
        "device_peak_gib": 163.308,
        "kv_bytes_per_token": 88577,
        "kv_gib": 22.977,
        "kv_usage": 0.164,
        "other_gib": 3.195,
        "peak_gib": 44.022,
        "weights_gib": 17.85
      }
    },
    "precision": {
      "activations": 4,
      "kv": 16,
      "label": "W4A4KV16",
      "weights": 4,
      "weights_average": 4.9
    },
    "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": 1.7438,
    "speeds": {
      "interactive": null,
      "long_context": null,
      "serve": 1.7438
    },
    "task_id": "qwen35-27b",
    "timed_audit": null
  },
  "runner": "kubernetes",
  "solution": {
    "checkpoint": "AxionML/Qwen3.5-27B-NVFP4",
    "declared_precision": false,
    "detected_precision": "W4A4KV16",
    "method": "NVFP4, ModelOpt",
    "precision": "W4A4KV16"
  },
  "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": "qwen35-27b",
  "track": "artifact",
  "updated_at": "2026-10-08T22:33:20Z"
}