Submission · qwen3-8b

8cf31177dcf0413a

by arena-check (participant) · dev split · runner kubernetes

stock-vLLM-plugin-arena-check RedHatAI/Qwen3-8B-FP8-dynamic W8A8KV8

done

Pipeline

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

Result

1.452LowBScore on the development split (never ranked)
W·A·KV W8A8KV8 declared · the service reads W8A8KV8
Accuracytier B · mean drop 0.78 points · worst domain long_context (upper bound 3.5)
Speed gain1.36× BF16 in the same pod (serve 1.44×, interactive 1.25×, long context 1.40×)
Memory gain1.78× BF16's peak (serve 1.85×, interactive 1.58×, long context 1.94×) · 30.1 GiB against 55.6 GiB on the first workload
E · Defficiency gain 1.56 · drift 0.69 points
Contractpassed, 8.00 bits per weight stored
GPU-hours0.60
Raw record
{
  "agent": "arena-check",
  "created_at": "2026-10-12T22:56:48Z",
  "error": null,
  "id": "8cf31177dcf0413a",
  "participant": "arena-check",
  "partition": "participant",
  "result": {
    "contract": {
      "activation_bits": 8,
      "bits_per_weight": 8.003226902173912,
      "foreign_tensors": {
        "bytes": 0,
        "count": 0,
        "largest": [],
        "share": 0.0
      },
      "format": "fp8-w8a8",
      "kv_bits": null,
      "model_bits_per_weight": 8.660807066158641,
      "ok": true,
      "quantization": {
        "activations": {
          "actorder": null,
          "block_structure": null,
          "dynamic": true,
          "group_size": null,
          "num_bits": 8,
          "observer": null,
          "observer_kwargs": {},
          "strategy": "token",
          "symmetric": true,
          "type": "float"
        },
        "format": "fp8-w8a8",
        "ignore": [
          "lm_head"
        ],
        "kv_quantized": false,
        "sparse": false,
        "weights": {
          "actorder": null,
          "block_structure": null,
          "dynamic": false,
          "group_size": null,
          "num_bits": 8,
          "observer": "mse",
          "observer_kwargs": {},
          "strategy": "channel",
          "symmetric": true,
          "type": "float"
        }
      },
      "rope_artifact": {
        "rope_theta": 1000000,
        "rope_type": "default"
      },
      "rope_source": {
        "rope_theta": 1000000,
        "rope_type": "default"
      },
      "safetensor_bytes": 9438581960,
      "sparse": false,
      "violations": [],
      "weight_bits": 8
    },
    "dev_quality": {
      "domains": {
        "code": {
          "bf16": 0.3380681818181818,
          "bf16_se": 0.047431027842082016,
          "candidate_lost": 0.0,
          "chance": 0.0,
          "drop": 0.005681818181818182,
          "headroom": 0.3380681818181818,
          "items": 88,
          "quantized": 0.33238636363636365,
          "recovery": 0.9831932773109244,
          "recovery_se": 0.02383649701391493,
          "se": 0.00805836120640874,
          "status": "valid",
          "upper": 0.01893782236636056
        },
        "commonsense": {
          "bf16": 0.7466666666666667,
          "bf16_se": 0.024134209165109787,
          "candidate_lost": 0.0,
          "chance": 0.24983333333333332,
          "drop": 0.0008333333333333334,
          "format_contract": {
            "bf16": 1.0,
            "quantized": 1.0
          },
          "headroom": 0.49683333333333335,
          "items": 300,
          "quantized": 0.7458333333333333,
          "recovery": 0.9983227104998322,
          "recovery_se": 0.023097242485515545,
          "se": 0.011475479974886973,
          "status": "valid",
          "upper": 0.019710497892022404
        },
        "expert": {
          "bf16": 0.44166666666666665,
          "bf16_se": 0.08491318532563476,
          "candidate_lost": 0.0,
          "chance": 0.25,
          "drop": -0.05,
          "format_contract": {
            "bf16": 0.9916666666666667,
            "quantized": 1.0
          },
          "headroom": 0.19166666666666665,
          "items": 30,
          "quantized": 0.49166666666666664,
          "recovery": 1.2608695652173914,
          "recovery_se": 0.24937180705779513,
          "se": 0.04779626301941073,
          "status": "R5_diagnostic",
          "upper": 0.028624852666930642
        },
        "knowledge": {
          "bf16": 0.44666666666666666,
          "bf16_se": 0.027939588170592746,
          "candidate_lost": 0.0,
          "chance": 0.10940343915343916,
          "drop": 0.006666666666666667,
          "format_contract": {
            "bf16": 0.92,
            "quantized": 0.9158333333333334
          },
          "headroom": 0.3372632275132275,
          "items": 300,
          "quantized": 0.44,
          "recovery": 0.9802330461111264,
          "recovery_se": 0.03880198605586273,
          "se": 0.013086483051123513,
          "status": "valid",
          "upper": 0.028193931285764848
        },
        "long_context": {
          "bf16": 0.8561666666666666,
          "bf16_se": 0.02867970594529983,
          "candidate_lost": 0.0,
          "chance": 0.0,
          "drop": 0.016833333333333336,
          "headroom": 0.8561666666666666,
          "items": 100,
          "quantized": 0.8393333333333334,
          "recovery": 0.9803387190967492,
          "recovery_se": 0.012753097837634363,
          "se": 0.010918777265321287,
          "status": "valid",
          "upper": 0.03479472193478685
        },
        "math": {
          "bf16": 0.89,
          "bf16_se": 0.019057569487578137,
          "candidate_lost": 0.0,
          "chance": 0.0,
          "drop": 0.00875,
          "headroom": 0.89,
          "items": 200,
          "quantized": 0.88125,
          "recovery": 0.9901685393258427,
          "recovery_se": 0.010418863724702894,
          "se": 0.009272788714985575,
          "status": "valid",
          "upper": 0.024003737436151272
        }
      },
      "evaluations": 1,
      "first_quality": null,
      "length": {
        "length_generations": 4544,
        "length_shift": 0.0023976825373244193,
        "length_shift_se": 0.007431886470118981,
        "length_tokens": 148.49559859154928,
        "length_tokens_bf16": 148.14040492957747
      },
      "recovery": 0.986451258468895,
      "recovery_se": 0.010730236806293977,
      "reference_rows": "/work/results/qwen-qwen3-8b/bf16-lb",
      "tier": "B",
      "valid_domains": 5,
      "worst_domain": "long_context",
      "worst_drop": 0.016833333333333336,
      "worst_upper": 0.03479472193478685
    },
    "dev_score": {
      "bf16_peak_gib": 55.593,
      "drift": 0.6894,
      "efficiency_gain": 1.5558,
      "mean_drop": 0.00775,
      "memories": {
        "interactive": 1.5843,
        "long_context": 1.9433,
        "serve": 1.8473
      },
      "memory": 1.785,
      "memory_saved": 0.4587,
      "peak_gib": 30.095,
      "precision": {
        "activations": 8,
        "kv": 8,
        "label": "W8A8KV8",
        "weights": 8,
        "weights_average": 8.0
      },
      "qualified": true,
      "reason": null,
      "score": 1.4521,
      "speed": 1.3592,
      "speeds": {
        "interactive": 1.2494,
        "long_context": 1.399,
        "serve": 1.4368
      }
    },
    "efficiency": {
      "interactive": 327.640349697288,
      "long_context": 1047.0827672609114,
      "serve": 2845.0435879059864
    },
    "engine": {
      "audit": null,
      "budget": {
        "interactive@1.02": {
          "budget_gib": 11.325,
          "cap_gib": 15.325,
          "factor": 1.02,
          "full_running": 1.0,
          "healthy": true,
          "running": 1.0,
          "value": 308.7506000661224
        },
        "long_context@1.02": {
          "budget_gib": 83.847,
          "cap_gib": 87.847,
          "factor": 1.02,
          "full_running": 64.0,
          "healthy": true,
          "running": 64.0,
          "value": 1045.9710146773723
        },
        "serve@1.02": {
          "budget_gib": 30.697,
          "cap_gib": 34.697,
          "factor": 1.02,
          "full_running": 32.0,
          "healthy": true,
          "running": 32.0,
          "value": 2789.918160879023
        }
      },
      "hard_cap": true,
      "startup": {
        "cuda_graph_gib": 0.44,
        "engine_init_seconds": 69.8,
        "kv_bytes_per_token": 73730.03096043378,
        "kv_cache_dtype": "fp8",
        "kv_cache_gib": 150.14,
        "kv_cache_tokens": 2186512.0,
        "load_seconds": 6.807922,
        "max_concurrency": {
          "sequences": 66.73,
          "tokens_per_request": 32768
        },
        "speculative_config": "None",
        "vllm_version": "0.1.dev18498+g7a33ba95f",
        "weights_gib": 8.84
      },
      "stored_gib": 8.79
    },
    "gpu_hours": 0.596,
    "kl": {
      "kl": 0.008560454064680034,
      "kl_median": 0.007647323705158094,
      "kl_per_token": 0.00866096821658328,
      "kl_se": 0.0013166281062096637,
      "sequences": 64,
      "tokens": 15418
    },
    "split": "dev",
    "workload": {
      "bf16": {
        "peak_gib": 55.593,
        "rate": 1980.141921938787
      },
      "submission": {
        "peak_gib": 30.095,
        "rate": 2845.0435879059864
      }
    }
  },
  "runner": "kubernetes",
  "solution": {
    "checkpoint": "RedHatAI/Qwen3-8B-FP8-dynamic",
    "declared_precision": true,
    "detected_precision": "W8A8KV8",
    "method": "stock-vLLM-plugin-arena-check",
    "precision": "W8A8KV8"
  },
  "split": "dev",
  "stages": [
    {
      "name": "build",
      "status": "done"
    },
    {
      "name": "contract",
      "status": "done"
    },
    {
      "name": "quality",
      "status": "done"
    },
    {
      "name": "efficiency",
      "status": "done"
    },
    {
      "name": "score",
      "status": "skipped"
    }
  ],
  "status": "done",
  "task_id": "qwen3-8b",
  "track": "engine",
  "updated_at": "2026-10-12T23:36:49Z"
}