Submission · qwen35-27b

`b2a9704d70cc480f`

by FP8 on the agent's coverage (reference) · test split · runner kubernetes

FP8 on the agent's coverage Qwen3.5-27B-FP8-GDN-calKV W8A8KV8

done

Pipeline

build done
contract done
quality done
efficiency done
score done

Result

pending LowBScore

speed or memory on the other workloads not measured

Entered as an engine that built its own checkpoint before the task fixed one; scored as the stock engine with the stock flags serves that checkpoint.

W·A·KV	W8A8KV8 declared · the service reads W8A8KV8
Accuracy	tier A · mean drop -0.10 points
Drift from BF16	KL 0.003 nats per token
Speed gain	1.30× BF16 in the same pod (serve 1.30×) (gate: at least 0.95× less its error)
Memory gain	— BF16's peak (serve 1.69×) · 45.9 GiB against BF16's 77.8 GiB on the first workload
Per-domain results on the hidden split are not released.

0.99 GPU-hours

Raw record

{
  "agent": "zoo-agentcov-fp8",
  "created_at": "2026-10-07T22:45:06Z",
  "error": null,
  "id": "b2a9704d70cc480f",
  "participant": "FP8 on the agent\u0027s coverage",
  "partition": "reference",
  "result": {
    "admitted": true,
    "anchor_efficiency": {
      "serve": 859.4183883619793
    },
    "bf16_peak_gib": 77.756,
    "efficiency": {
      "serve": 1119.6681387470678
    },
    "gpu_hours": 0.987,
    "kl": {
      "kl": 0.0025021658872059374,
      "kl_median": 0.002487508584996842,
      "kl_per_token": 0.0025206417619092914,
      "kl_se": 0.0005904371932053071,
      "sequences": 64,
      "tokens": 16088
    },
    "mean_drop": -0.00105,
    "memories": {
      "interactive": null,
      "long_context": null,
      "serve": 1.6942
    },
    "memory": null,
    "memory_saved": 0.4097,
    "metric": "lowbit_score",
    "peak_gib": 45.896,
    "peak_memory": {
      "serve": {
        "device_peak_gib": 161.913,
        "kv_bytes_per_token": 53165,
        "kv_gib": 13.791,
        "kv_usage": 0.1069,
        "other_gib": 4.425,
        "peak_gib": 45.896,
        "weights_gib": 27.68
      }
    },
    "precision": {
      "activations": 8,
      "kv": 8,
      "label": "W8A8KV8",
      "weights": 8,
      "weights_average": 8.02
    },
    "qualified": false,
    "quality": {
      "tier": "A",
      "valid_domains": 5
    },
    "reason": "speed or memory on the other workloads not measured",
    "score": null,
    "score_se": null,
    "served_as": "checkpoint",
    "speed": null,
    "speed_serve": 1.3028,
    "speeds": {
      "interactive": null,
      "long_context": null,
      "serve": 1.3028
    },
    "task_id": "qwen35-27b",
    "timed_audit": null
  },
  "runner": "kubernetes",
  "solution": {
    "checkpoint": "Qwen3.5-27B-FP8-GDN-calKV",
    "declared_precision": true,
    "detected_precision": "W8A8KV8",
    "method": "FP8 on the agent\u0027s coverage",
    "precision": "W8A8KV8"
  },
  "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"
}