Qwen3-8B · scored

stock engine (A/A)

reference other judged submitted 10-24 03:15 UTC

Passed every gate.

SpecScore1.001geometric mean of the three speedups
solo1.00× one lane; passes 1.00, 1.00, 0.99
team0.99× team lanes busy; passes 1.02, 0.99, 0.99
busy1.01× KV cache nearly full; passes 1.01, 0.99, 1.02
CheckMeasuredLimit
device memory162.2 GiB≤ 165.2 GiB pass
fixed texts' log-probabilities0≤ 0.001 pass
sampled replies' likelihood+0.0634 ± 0.0809within ±0.02 at 95% pass
share of first choices-0.0147 ± 0.0147within ±0.003 at 95% pass
greedy agreement1.000≥ 0.5 pass
reply lengths0.999 [0.918, 1.081]1 ± 0.05 at 95% pass

Speculation

WorkloadTokens / verify stepSpeedup
solo– 1.00
team– 0.99
busy– 1.01

Speculator: none
Engine: uploaded with the submission

Everything the judge recorded
{
 "ratios": {
  "solo": {
   "passes": [
    1.0000385424846772,
    1.0012005305030602,
    0.9904126634610351
   ],
   "ratio": 1.0000385424846772
  },
  "team": {
   "passes": [
    1.0186178038684752,
    0.9913053896047394,
    0.9876315374651222
   ],
   "ratio": 0.9913053896047394
  },
  "busy": {
   "passes": [
    1.0110412656856833,
    0.9935443320688605,
    1.0245015035320995
   ],
   "ratio": 1.0110412656856833
  }
 },
 "specscore": 1.000762513431338,
 "greedy_agreement": 1.0,
 "likelihood": {
  "gap": 0.06343148946821608,
  "se": 0.04128231504049546,
  "candidate": -0.6292665117843087,
  "stock": -0.6926980012525248
 },
 "argmax": {
  "gap": -0.014675071204507506,
  "se": 0.007506715210185353,
  "candidate": 0.13409229079641505,
  "stock": 0.14876736200092255
 },
 "fidelity": 0.0,
 "lengths": {
  "solo": {
   "ratio": 1.0,
   "low": 0.6850711091871367,
   "high": 1.3149288908128633,
   "paired": true,
   "replies": 240
  },
  "team": {
   "ratio": 0.9975241396385244,
   "low": 0.8562298990093652,
   "high": 1.1388183802676837,
   "paired": true,
   "replies": 1257
  },
  "busy": {
   "ratio": 1.0003583838446795,
   "low": 0.895397638304234,
   "high": 1.105319129385125,
   "paired": true,
   "replies": 1815
  },
  "pooled": {
   "ratio": 0.9993341795419988,
   "low": 0.9177464988682371,
   "high": 1.0809218602157602,
   "paired": true,
   "replies": 3312
  }
 },
 "memory": {
  "ok": true,
  "peak_mib": 166072.0,
  "allowed_mib": 169119.1
 },
 "acceptance": {},
 "limits": {
  "greedy_agreement_floor": 0.5,
  "likelihood_bound": 0.02,
  "argmax_bound": 0.003,
  "fidelity_ceiling": 0.001,
  "length_margin": 0.05,
  "memory_slack_mib": 4096
 },
 "reason": null,
 "score": 1.000762513431338
}