Qwen3.5-27B · scored

MTP k=1

reference mtp judged submitted 10-23 11:15 UTC

Passed every gate.

SpecScore1.108geometric mean of the three speedups

solo1.29× one lane; passes 1.25, 1.29, 1.34

team1.18× team lanes busy; passes 1.11, 1.18, 1.20

busy0.89× KV cache nearly full; passes 0.97, 0.89, 0.83

Lossless gate

what each check is →

Check	Measured	Limit
device memory	161.3 GiB	≤ 165.2 GiB	pass
fixed texts' log-probabilities	0.012	≤ 0.02	pass
sampled replies' likelihood	+0.0025 ± 0.0502	within ±0.05 at 95%	pass
share of first choices	-0.0010 ± 0.0117	within ±0.01 at 95%	pass
greedy agreement	0.515	none on this task	n/a
reply lengths	–	1 ± 0.05 at 95%	pass

Speculation

Workload	Tokens / verify step	Speedup
solo	1.88	1.29
team	1.89	1.18
busy	1.85	0.89

Speculator: none
Engine: uploaded with the submission

Everything the judge recorded

{
 "ratios": {
  "solo": {
   "passes": [
    1.251423706826688,
    1.294871155443465,
    1.3378688652967927
   ],
   "ratio": 1.294871155443465
  },
  "team": {
   "passes": [
    1.111305711295507,
    1.181251282273152,
    1.204029121156538
   ],
   "ratio": 1.181251282273152
  },
  "busy": {
   "passes": [
    0.9721688257009765,
    0.889813411278546,
    0.8290520898549844
   ],
   "ratio": 0.889813411278546
  }
 },
 "specscore": 1.1082113631145032,
 "greedy_agreement": 0.5150070440243575,
 "likelihood": {
  "gap": 0.002452181568941003,
  "se": 0.02559000206175653,
  "candidate": -0.47145227485054614,
  "stock": -0.47390445641948714
 },
 "argmax": {
  "gap": -0.0010319330208626576,
  "se": 0.0059525668606879995,
  "candidate": 0.12913646756083788,
  "stock": 0.13016840058170054
 },
 "fidelity": 0.011885751759680816,
 "lengths": {
  "solo": {
   "ratio": 0.8628201065689641,
   "low": 0.5193543533368512,
   "high": 1.206285859801077
  },
  "team": {
   "ratio": 0.9643550519908614,
   "low": 0.6923727849828659,
   "high": 1.2363373189988567
  },
  "busy": {
   "ratio": 0.9044800712728782,
   "low": 0.6847508040698181,
   "high": 1.1242093384759384
  }
 },
 "memory": {
  "ok": true,
  "peak_mib": 165134.0,
  "allowed_mib": 169119.1
 },
 "acceptance": {
  "solo": 1.8791427859456766,
  "team": 1.8857501698614343,
  "busy": 1.8534476808393152
 },
 "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": 1.1082113631145032
}