Qwen3.5-27B · scored

stock engine (A/A)

reference other judged submitted 10-23 08:29 UTC

Passed every gate.

SpecScore1.020geometric mean of the three speedups

solo0.98× one lane; passes 0.96, 0.98, 1.00

team1.03× team lanes busy; passes 1.11, 1.03, 0.92

busy1.05× KV cache nearly full; passes 1.05, 1.06, 1.01

Lossless gate

what each check is →

Check	Measured	Limit
device memory	162.5 GiB	≤ 165.2 GiB	pass
fixed texts' log-probabilities	0.0078	≤ 0.02	pass
sampled replies' likelihood	-0.0117 ± 0.0528	within ±0.05 at 95%	pass
share of first choices	+0.0036 ± 0.0120	within ±0.01 at 95%	pass
greedy agreement	0.763	none on this task	n/a
reply lengths	–	1 ± 0.05 at 95%	pass

Speculation

Workload	Tokens / verify step	Speedup
solo	–	0.98
team	–	1.03
busy	–	1.05

Speculator: none
Engine: uploaded with the submission

Everything the judge recorded

{
 "ratios": {
  "solo": {
   "passes": [
    0.9608282011567039,
    0.9826434921591861,
    1.0038542588246389
   ],
   "ratio": 0.9826434921591861
  },
  "team": {
   "passes": [
    1.1063111185890517,
    1.0340405551888396,
    0.9207437571768063
   ],
   "ratio": 1.0340405551888396
  },
  "busy": {
   "passes": [
    1.0458767065312737,
    1.0568695892290394,
    1.0148011809600916
   ],
   "ratio": 1.0458767065312737
  }
 },
 "specscore": 1.0204804326352712,
 "greedy_agreement": 0.7634347175262509,
 "likelihood": {
  "gap": -0.011663239906921807,
  "se": 0.0269207274647701,
  "candidate": -0.4907981348050229,
  "stock": -0.47913489489810107
 },
 "argmax": {
  "gap": 0.003565088521198967,
  "se": 0.006097526546928617,
  "candidate": 0.13303075082682428,
  "stock": 0.1294656623056253
 },
 "fidelity": 0.007776617449717978,
 "lengths": {
  "solo": {
   "ratio": 0.935739685936369,
   "low": 0.5149504707311471,
   "high": 1.3565289011415909
  },
  "team": {
   "ratio": 1.0267405181587816,
   "low": 0.7268822134490345,
   "high": 1.3265988228685288
  },
  "busy": {
   "ratio": 1.0343258673569764,
   "low": 0.7949103007660526,
   "high": 1.2737414339479003
  }
 },
 "memory": {
  "ok": true,
  "peak_mib": 166356.0,
  "allowed_mib": 169119.1
 },
 "acceptance": {},
 "limits": {
  "greedy_agreement_floor": null,
  "likelihood_bound": 0.05,
  "argmax_bound": 0.01,
  "fidelity_ceiling": 0.02,
  "length_margin": 0.05,
  "memory_slack_mib": 4096
 },
 "reason": null,
 "score": 1.0204804326352712
}