Qwen3.5-27B · scored

stock engine (A/A)

reference other judged submitted 10-23 12:25 UTC

Passed every gate.

SpecScore0.995geometric mean of the three speedups

solo1.00× one lane; passes 1.00, 1.00, 1.02

team1.00× team lanes busy; passes 1.02, 1.00, 0.89

busy0.98× KV cache nearly full; passes 1.09, 0.89, 0.98

Lossless gate

what each check is →

Check	Measured	Limit
device memory	163.3 GiB	≤ 165.2 GiB	pass
fixed texts' log-probabilities	1.2e-07	≤ 0.017	pass
sampled replies' likelihood	+0.0153 ± 0.0501	within ±0.02 at 95%	pass
share of first choices	-0.0021 ± 0.0114	within ±0.003 at 95%	pass
greedy agreement	1.000	none on this task	n/a
reply lengths	–	1 ± 0.05 at 95%	pass

Speculation

Workload	Tokens / verify step	Speedup
solo	–	1.00
team	–	1.00
busy	–	0.98

Speculator: none
Engine: uploaded with the submission

Everything the judge recorded

{
 "ratios": {
  "solo": {
   "passes": [
    1.0026712304330845,
    1.004451786528974,
    1.0234452332351283
   ],
   "ratio": 1.004451786528974
  },
  "team": {
   "passes": [
    1.0220941372948782,
    0.998162103749363,
    0.8871137016942595
   ],
   "ratio": 0.998162103749363
  },
  "busy": {
   "passes": [
    1.0866953556039196,
    0.8868853930499235,
    0.9837399479976354
   ],
   "ratio": 0.9837399479976354
  }
 },
 "specscore": 0.9954134246220057,
 "greedy_agreement": 1.0,
 "likelihood": {
  "gap": 0.015265904876172243,
  "se": 0.025568861305435828,
  "candidate": -0.48197303389100965,
  "stock": -0.4972389387671819
 },
 "argmax": {
  "gap": -0.0021309659699700334,
  "se": 0.005828872455698585,
  "candidate": 0.13061049911838132,
  "stock": 0.13274146508835136
 },
 "fidelity": 1.2405465865759965e-07,
 "lengths": {
  "solo": {
   "ratio": 1.0,
   "low": 0.5731882183056475,
   "high": 1.4268117816943526
  },
  "team": {
   "ratio": 0.9623778185926303,
   "low": 0.6776160794291143,
   "high": 1.2471395577561464
  },
  "busy": {
   "ratio": 1.0530163813220934,
   "low": 0.7874178581650114,
   "high": 1.3186149044791753
  }
 },
 "memory": {
  "ok": true,
  "peak_mib": 167172.0,
  "allowed_mib": 169119.1
 },
 "acceptance": {},
 "limits": {
  "greedy_agreement_floor": null,
  "likelihood_bound": 0.02,
  "argmax_bound": 0.003,
  "fidelity_ceiling": 0.017,
  "length_margin": 0.05,
  "memory_slack_mib": 4096
 },
 "reason": null,
 "score": 0.9954134246220057
}