Jev / Luna
How much quality, how quickly, at what cost. Identical output schemas; separately scored tasks with checkable answers.
Quality and response-time tables use valid measured responses. “Accuracy · all” also counts invalid requests as failures. Cost includes all measured requests. A dash means unavailable; unknown cost is never assumed to be zero.
Reading comprehension · BoolQ
Accuracy: higher is better. Brier and log loss: lower is better.
| Model | Valid / requests | Accuracy · valid | Accuracy · all | Brier | Log loss | p50 / p95 | USD / 1,000 | USD / correct |
|---|---|---|---|---|---|---|---|---|
| Jev | 60 / 60 | 95.0% | 95.0% | 0.068 | 0.248 | 0.236 / 0.372 s | $0.019847 | $0.000021 |
| Luna | 60 / 60 | 81.7% | 81.7% | 0.168 | 3.676 | 1.148 / 1.470 s | $0.044483 | $0.000054 |
Median complete-response time
Valid responses only. Smaller is faster.
Paired comparison · 95% intervals
Luna − Jev accuracy-13.33 pp [-21.67, -5.00]
Luna / Jev median time4.86× [4.42, 5.10]
60 of 60 cases paired; 0 excluded. Positive accuracy difference favours Luna. A time ratio below 1 favours Luna.Billing and routing detail
- Jev: $0.001191 known measured billing · 0 missing bills · resolved model typesafe/jev-1.13-20260917 · provider TypeSafe
- Luna: $0.002669 known measured billing · 0 missing bills · resolved model openai/gpt-6-luna · provider OpenAI
All-request timing, additional paired metrics, interval seeds, and exact unrounded numbers are in summary.json. Billing includes invalid measured requests; warmups are shown separately above.
Policy decisions
Accuracy: higher is better. Brier and log loss: lower is better.
| Model | Valid / requests | Accuracy · valid | Accuracy · all | Brier | Log loss | p50 / p95 | USD / 1,000 | USD / correct |
|---|---|---|---|---|---|---|---|---|
| Jev | 60 / 60 | 100.0% | 100.0% | 0.008 | 0.067 | 0.243 / 0.331 s | $0.020927 | $0.000021 |
| Luna | 60 / 60 | 100.0% | 100.0% | 0.000 | 0.000 | 1.149 / 1.647 s | $0.045735 | $0.000046 |
Median complete-response time
Valid responses only. Smaller is faster.
Paired comparison · 95% intervals
Luna − Jev accuracy+0.00 pp [0.00, 0.00]
Luna / Jev median time4.72× [4.26, 5.05]
60 of 60 cases paired; 0 excluded. Positive accuracy difference favours Luna. A time ratio below 1 favours Luna.Billing and routing detail
- Jev: $0.001256 known measured billing · 0 missing bills · resolved model typesafe/jev-1.13-20260917 · provider TypeSafe
- Luna: $0.002744 known measured billing · 0 missing bills · resolved model openai/gpt-6-luna · provider OpenAI
All-request timing, additional paired metrics, interval seeds, and exact unrounded numbers are in summary.json. Billing includes invalid measured requests; warmups are shown separately above.
Exact probabilities
MAE, RMSE, and excess Brier: lower is better. Expected Brier includes irreducible outcome uncertainty.
| Model | Valid / requests | MAE | RMSE | Excess Brier | Expected Brier | p50 / p95 | USD / 1,000 |
|---|---|---|---|---|---|---|---|
| Jev | 60 / 60 | 0.064 | 0.100 | 0.010 | 0.156 | 0.237 / 0.311 s | $0.018645 |
| Luna | 60 / 60 | 0.033 | 0.074 | 0.006 | 0.152 | 1.143 / 1.596 s | $0.041408 |
Median complete-response time
Valid responses only. Smaller is faster.
Paired comparison · 95% intervals
Luna − Jev mae-0.03 [-0.05, -0.01]
Luna / Jev median time4.81× [4.47, 5.37]
60 of 60 cases paired; 0 excluded. Negative MAE difference favours Luna. A time ratio below 1 favours Luna.Billing and routing detail
- Jev: $0.001119 known measured billing · 0 missing bills · resolved model typesafe/jev-1.13-20260917 · provider TypeSafe
- Luna: $0.002485 known measured billing · 0 missing bills · resolved model openai/gpt-6-luna · provider OpenAI
All-request timing, additional paired metrics, interval seeds, and exact unrounded numbers are in summary.json. Billing includes invalid measured requests; warmups are shown separately above.
Read the result in context
- This is a small benchmark. Intervals describe variation across the sampled cases; they do not prove a general model ranking.
- Luna supplies generated numerical probabilities; Jev supplies its native Noul probabilities. Neither adapter extracts next-token probabilities. Brier combines calibration and discrimination; this small sample cannot establish general calibration.
- Each task has a separate quality scale. Quality is not pooled across tasks; no overall winner is inferred.
- BoolQ is public and may have appeared in training data. Reading-comprehension results can be affected by contamination.
- Paired comparisons include only cases with valid responses from both models. Check success rates and excluded pairs before comparing conditional quality.
- Latency includes the OpenRouter/provider/network path. Provider routing, caching, load, reasoning settings, and output length can change speed and price.
- The intended primary protocol is Luna reasoning=none, sequential randomized interleaving, and no retries. The recorded run metadata is the source of truth for whether that protocol was used.
- 95% intervals are exploratory percentile bootstrap intervals without multiple-comparison correction. Zero-width intervals can occur in a small or uniform sample.
- Costs are USD as reported by the runner. Missing billing remains unknown; cost per 1,000 is a sample-based projection, not a price guarantee.
Every request
Inspect the record behind each aggregate. Includes warmups and failed requests.
364 requests shown
| Case | Experiment | Model | Phase | Target | p(yes) | Result | Seconds | USD | Error |
|---|---|---|---|---|---|---|---|---|---|
| warmup-yes | warmup | Luna | warmup | 1.0000 | 1.0000 | Valid | 1.134 | $0.000030 | |
| warmup-yes | warmup | Jev | warmup | 1.0000 | 0.9600 | Valid | 0.241 | $0.000014 | |
| warmup-no | warmup | Jev | warmup | 0.0000 | 0.0400 | Valid | 0.251 | $0.000014 | |
| warmup-no | warmup | Luna | warmup | 0.0000 | 0.0000 | Valid | 1.973 | $0.000030 | |
| policy-016 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 0.900 | $0.000046 | |
| policy-016 | policy | Jev | measured | 0.0000 | 0.0600 | Valid | 0.238 | $0.000021 | |
| probability-044 | probability | Jev | measured | 0.0667 | 0.1900 | Valid | 0.219 | $0.000017 | |
| probability-044 | probability | Luna | measured | 0.0667 | 0.0667 | Valid | 1.040 | $0.000041 | |
| boolq-dev-1550 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.329 | $0.000055 | |
| boolq-dev-1550 | boolq | Jev | measured | 0.0000 | 0.0600 | Valid | 0.347 | $0.000024 | |
| boolq-dev-0579 | boolq | Jev | measured | 1.0000 | 0.6300 | Valid | 0.240 | $0.000018 | |
| boolq-dev-0579 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.316 | $0.000040 | |
| boolq-dev-1066 | boolq | Jev | measured | 1.0000 | 0.8600 | Valid | 0.231 | $0.000020 | |
| boolq-dev-1066 | boolq | Luna | measured | 1.0000 | 0.9900 | Valid | 0.899 | $0.000044 | |
| policy-021 | policy | Jev | measured | 0.0000 | 0.0400 | Valid | 0.222 | $0.000021 | |
| policy-021 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 0.945 | $0.000046 | |
| boolq-dev-1114 | boolq | Jev | measured | 1.0000 | 0.9900 | Valid | 0.224 | $0.000022 | |
| boolq-dev-1114 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.075 | $0.000048 | |
| boolq-dev-3114 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.467 | $0.000042 | |
| boolq-dev-3114 | boolq | Jev | measured | 1.0000 | 0.9900 | Valid | 0.246 | $0.000019 | |
| probability-030 | probability | Jev | measured | 0.4583 | 0.4100 | Valid | 0.219 | $0.000020 | |
| probability-030 | probability | Luna | measured | 0.4583 | 0.4167 | Valid | 1.065 | $0.000045 | |
| probability-007 | probability | Luna | measured | 0.7255 | 0.8214 | Valid | 1.260 | $0.000045 | |
| probability-007 | probability | Jev | measured | 0.7255 | 0.4900 | Valid | 0.244 | $0.000020 | |
| probability-011 | probability | Jev | measured | 0.5415 | 0.5900 | Valid | 0.227 | $0.000020 | |
| probability-011 | probability | Luna | measured | 0.5415 | 0.5840 | Valid | 1.412 | $0.000044 | |
| probability-025 | probability | Luna | measured | 1.0000 | 1.0000 | Valid | 1.061 | $0.000039 | |
| probability-025 | probability | Jev | measured | 1.0000 | 0.9900 | Valid | 0.223 | $0.000018 | |
| boolq-dev-0767 | boolq | Luna | measured | 1.0000 | 0.5000 | Valid | 0.987 | $0.000048 | |
| boolq-dev-0767 | boolq | Jev | measured | 1.0000 | 0.6200 | Valid | 0.261 | $0.000022 | |
| probability-057 | probability | Luna | measured | 0.5500 | 0.5500 | Valid | 0.988 | $0.000040 | |
| probability-057 | probability | Jev | measured | 0.5500 | 0.5600 | Valid | 0.225 | $0.000018 | |
| boolq-dev-1447 | boolq | Jev | measured | 1.0000 | 0.9700 | Valid | 0.226 | $0.000018 | |
| boolq-dev-1447 | boolq | Luna | measured | 1.0000 | 0.0000 | Valid | 1.260 | $0.000041 | |
| boolq-dev-2791 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 0.908 | $0.000040 | |
| boolq-dev-2791 | boolq | Jev | measured | 0.0000 | 0.0500 | Valid | 0.230 | $0.000018 | |
| probability-031 | probability | Luna | measured | 0.5357 | 0.5357 | Valid | 1.079 | $0.000041 | |
| probability-031 | probability | Jev | measured | 0.5357 | 0.5200 | Valid | 0.346 | $0.000018 | |
| probability-033 | probability | Jev | measured | 0.4421 | 0.5600 | Valid | 0.256 | $0.000018 | |
| probability-033 | probability | Luna | measured | 0.4421 | 0.4421 | Valid | 1.005 | $0.000041 | |
| boolq-dev-2493 | boolq | Jev | measured | 1.0000 | 0.9600 | Valid | 0.268 | $0.000018 | |
| boolq-dev-2493 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.180 | $0.000041 | |
| boolq-dev-1570 | boolq | Jev | measured | 0.0000 | 0.0300 | Valid | 0.236 | $0.000022 | |
| boolq-dev-1570 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.027 | $0.000048 | |
| policy-007 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.143 | $0.000046 | |
| policy-007 | policy | Jev | measured | 0.0000 | 0.2300 | Valid | 0.231 | $0.000021 | |
| boolq-dev-1979 | boolq | Jev | measured | 1.0000 | 0.9900 | Valid | 0.221 | $0.000020 | |
| boolq-dev-1979 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.352 | $0.000045 | |
| probability-018 | probability | Jev | measured | 0.7317 | 0.7700 | Valid | 0.224 | $0.000018 | |
| probability-018 | probability | Luna | measured | 0.7317 | 0.7585 | Valid | 1.445 | $0.000040 | |
| boolq-dev-0055 | boolq | Jev | measured | 0.0000 | 0.3800 | Valid | 0.252 | $0.000044 | |
| boolq-dev-0055 | boolq | Luna | measured | 0.0000 | 0.1500 | Valid | 1.038 | $0.000103 | |
| boolq-dev-2504 | boolq | Jev | measured | 0.0000 | 0.2300 | Valid | 0.242 | $0.000017 | |
| boolq-dev-2504 | boolq | Luna | measured | 0.0000 | 0.9800 | Valid | 1.050 | $0.000040 | |
| probability-039 | probability | Jev | measured | 0.2000 | 0.2300 | Valid | 0.222 | $0.000018 | |
| probability-039 | probability | Luna | measured | 0.2000 | 0.2000 | Valid | 1.079 | $0.000040 | |
| probability-055 | probability | Jev | measured | 0.7051 | 0.7200 | Valid | 0.262 | $0.000017 | |
| probability-055 | probability | Luna | measured | 0.7051 | 0.7051 | Valid | 1.041 | $0.000040 | |
| policy-013 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.017 | $0.000045 | |
| policy-013 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.222 | $0.000020 | |
| policy-012 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.234 | $0.000021 | |
| policy-012 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.270 | $0.000046 | |
| policy-025 | policy | Jev | measured | 0.0000 | 0.0700 | Valid | 0.254 | $0.000021 | |
| policy-025 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.173 | $0.000046 | |
| policy-047 | policy | Jev | measured | 0.0000 | 0.0400 | Valid | 0.254 | $0.000021 | |
| policy-047 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 0.982 | $0.000046 | |
| policy-041 | policy | Jev | measured | 0.0000 | 0.2300 | Valid | 0.256 | $0.000021 | |
| policy-041 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 0.990 | $0.000046 | |
| boolq-dev-0448 | boolq | Jev | measured | 1.0000 | 0.6100 | Valid | 0.221 | $0.000019 | |
| boolq-dev-0448 | boolq | Luna | measured | 1.0000 | 0.0000 | Valid | 0.978 | $0.000043 | |
| probability-042 | probability | Luna | measured | 0.0000 | 0.0000 | Valid | 1.276 | $0.000038 | |
| probability-042 | probability | Jev | measured | 0.0000 | 0.0300 | Valid | 0.229 | $0.000018 | |
| policy-039 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.226 | $0.000021 | |
| policy-039 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.592 | $0.000046 | |
| policy-036 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.003 | $0.000045 | |
| policy-036 | policy | Jev | measured | 0.0000 | 0.0900 | Valid | 0.263 | $0.000020 | |
| boolq-dev-1625 | boolq | Jev | measured | 1.0000 | 0.6600 | Valid | 0.222 | $0.000017 | |
| boolq-dev-1625 | boolq | Luna | measured | 1.0000 | 0.2500 | Valid | 1.260 | $0.000038 | |
| policy-040 | policy | Jev | measured | 1.0000 | 0.9700 | Valid | 0.229 | $0.000021 | |
| policy-040 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 0.951 | $0.000046 | |
| policy-001 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.279 | $0.000021 | |
| policy-001 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.065 | $0.000046 | |
| policy-028 | policy | Jev | measured | 0.0000 | 0.2000 | Valid | 0.400 | $0.000021 | |
| policy-028 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.499 | $0.000046 | |
| probability-014 | probability | Jev | measured | 0.9500 | 0.9300 | Valid | 0.253 | $0.000018 | |
| probability-014 | probability | Luna | measured | 0.9500 | 0.9500 | Valid | 1.437 | $0.000040 | |
| boolq-dev-3009 | boolq | Jev | measured | 1.0000 | 0.9400 | Valid | 0.294 | $0.000019 | |
| boolq-dev-3009 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.035 | $0.000041 | |
| boolq-dev-0341 | boolq | Jev | measured | 0.0000 | 0.0300 | Valid | 0.233 | $0.000021 | |
| boolq-dev-0341 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.478 | $0.000047 | |
| probability-020 | probability | Jev | measured | 0.5459 | 0.3800 | Valid | 0.245 | $0.000020 | |
| probability-020 | probability | Luna | measured | 0.5459 | 0.4000 | Valid | 0.972 | $0.000044 | |
| probability-040 | probability | Luna | measured | 0.8317 | 0.5470 | Valid | 1.235 | $0.000044 | |
| probability-040 | probability | Jev | measured | 0.8317 | 0.7200 | Valid | 0.239 | $0.000020 | |
| boolq-dev-3165 | boolq | Luna | measured | 1.0000 | 0.5000 | Valid | 1.256 | $0.000036 | |
| boolq-dev-3165 | boolq | Jev | measured | 1.0000 | 0.6300 | Valid | 0.298 | $0.000016 | |
| probability-024 | probability | Luna | measured | 0.6759 | 0.5500 | Valid | 2.750 | $0.000044 | |
| probability-024 | probability | Jev | measured | 0.6759 | 0.5000 | Valid | 0.248 | $0.000020 | |
| boolq-dev-1171 | boolq | Jev | measured | 0.0000 | 0.2600 | Valid | 0.227 | $0.000017 | |
| boolq-dev-1171 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.331 | $0.000038 | |
| probability-028 | probability | Jev | measured | 0.8260 | 0.6800 | Valid | 0.226 | $0.000020 | |
| probability-028 | probability | Luna | measured | 0.8260 | 0.8372 | Valid | 0.999 | $0.000044 | |
| policy-027 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.063 | $0.000045 | |
| policy-027 | policy | Jev | measured | 0.0000 | 0.0900 | Valid | 0.228 | $0.000020 | |
| policy-015 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.246 | $0.000020 | |
| policy-015 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.016 | $0.000045 | |
| probability-017 | probability | Luna | measured | 0.5167 | 0.5167 | Valid | 1.451 | $0.000042 | |
| probability-017 | probability | Jev | measured | 0.5167 | 0.5600 | Valid | 0.220 | $0.000018 | |
| boolq-dev-2552 | boolq | Jev | measured | 1.0000 | 0.9900 | Valid | 0.259 | $0.000019 | |
| boolq-dev-2552 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.140 | $0.000042 | |
| probability-052 | probability | Luna | measured | 0.4840 | 0.5247 | Valid | 1.052 | $0.000045 | |
| probability-052 | probability | Jev | measured | 0.4840 | 0.5500 | Valid | 0.236 | $0.000020 | |
| probability-056 | probability | Jev | measured | 0.7537 | 0.7700 | Valid | 0.221 | $0.000020 | |
| probability-056 | probability | Luna | measured | 0.7537 | 0.7283 | Valid | 1.880 | $0.000044 | |
| probability-032 | probability | Jev | measured | 0.2000 | 0.2200 | Valid | 0.234 | $0.000018 | |
| probability-032 | probability | Luna | measured | 0.2000 | 0.2000 | Valid | 0.939 | $0.000040 | |
| boolq-dev-3125 | boolq | Jev | measured | 1.0000 | 0.9700 | Valid | 0.216 | $0.000022 | |
| boolq-dev-3125 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.294 | $0.000048 | |
| boolq-dev-2098 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.501 | $0.000043 | |
| boolq-dev-2098 | boolq | Jev | measured | 1.0000 | 0.9500 | Valid | 0.228 | $0.000019 | |
| policy-046 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.111 | $0.000045 | |
| policy-046 | policy | Jev | measured | 0.0000 | 0.0700 | Valid | 0.264 | $0.000020 | |
| probability-048 | probability | Luna | measured | 0.0778 | 0.0778 | Valid | 0.945 | $0.000045 | |
| probability-048 | probability | Jev | measured | 0.0778 | 0.1700 | Valid | 0.322 | $0.000020 | |
| policy-002 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.228 | $0.000021 | |
| policy-002 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.478 | $0.000046 | |
| policy-056 | policy | Jev | measured | 0.0000 | 0.0900 | Valid | 0.242 | $0.000021 | |
| policy-056 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 0.969 | $0.000046 | |
| policy-043 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.419 | $0.000045 | |
| policy-043 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.334 | $0.000020 | |
| probability-005 | probability | Jev | measured | 0.0250 | 0.0600 | Valid | 0.248 | $0.000018 | |
| probability-005 | probability | Luna | measured | 0.0250 | 0.0250 | Valid | 1.301 | $0.000040 | |
| probability-021 | probability | Jev | measured | 0.2000 | 0.2200 | Valid | 0.254 | $0.000018 | |
| probability-021 | probability | Luna | measured | 0.2000 | 0.2000 | Valid | 1.372 | $0.000040 | |
| boolq-dev-1517 | boolq | Jev | measured | 0.0000 | 0.8600 | Valid | 0.236 | $0.000019 | |
| boolq-dev-1517 | boolq | Luna | measured | 0.0000 | 1.0000 | Valid | 1.294 | $0.000043 | |
| probability-023 | probability | Jev | measured | 0.1978 | 0.5800 | Valid | 0.224 | $0.000020 | |
| probability-023 | probability | Luna | measured | 0.1978 | 0.3600 | Valid | 1.014 | $0.000044 | |
| probability-022 | probability | Luna | measured | 0.6976 | 0.4446 | Valid | 1.017 | $0.000044 | |
| probability-022 | probability | Jev | measured | 0.6976 | 0.6400 | Valid | 0.237 | $0.000020 | |
| boolq-dev-1859 | boolq | Jev | measured | 0.0000 | 0.1700 | Valid | 0.237 | $0.000019 | |
| boolq-dev-1859 | boolq | Luna | measured | 0.0000 | 0.0500 | Valid | 1.245 | $0.000045 | |
| probability-027 | probability | Luna | measured | 0.2333 | 0.2333 | Valid | 1.102 | $0.000041 | |
| probability-027 | probability | Jev | measured | 0.2333 | 0.3600 | Valid | 0.237 | $0.000018 | |
| policy-049 | policy | Jev | measured | 1.0000 | 0.9900 | Valid | 0.262 | $0.000020 | |
| policy-049 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 0.953 | $0.000045 | |
| policy-034 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.223 | $0.000021 | |
| policy-034 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.015 | $0.000046 | |
| boolq-dev-1335 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 0.954 | $0.000057 | |
| boolq-dev-1335 | boolq | Jev | measured | 1.0000 | 0.9200 | Valid | 0.388 | $0.000025 | |
| boolq-dev-1092 | boolq | Jev | measured | 0.0000 | 0.0500 | Valid | 0.228 | $0.000020 | |
| boolq-dev-1092 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.234 | $0.000046 | |
| policy-023 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.255 | $0.000020 | |
| policy-023 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.302 | $0.000045 | |
| boolq-dev-2041 | boolq | Jev | measured | 0.0000 | 0.0300 | Valid | 0.236 | $0.000021 | |
| boolq-dev-2041 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.064 | $0.000047 | |
| probability-038 | probability | Luna | measured | 0.0000 | 0.0000 | Valid | 0.981 | $0.000037 | |
| probability-038 | probability | Jev | measured | 0.0000 | 0.0500 | Valid | 0.227 | $0.000017 | |
| policy-057 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.272 | $0.000046 | |
| policy-057 | policy | Jev | measured | 0.0000 | 0.0700 | Valid | 0.226 | $0.000021 | |
| probability-016 | probability | Luna | measured | 0.0237 | 0.0237 | Valid | 1.096 | $0.000041 | |
| probability-016 | probability | Jev | measured | 0.0237 | 0.1500 | Valid | 0.234 | $0.000017 | |
| boolq-dev-0525 | boolq | Jev | measured | 0.0000 | 0.0600 | Valid | 0.235 | $0.000016 | |
| boolq-dev-0525 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.016 | $0.000036 | |
| boolq-dev-0980 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.169 | $0.000038 | |
| boolq-dev-0980 | boolq | Jev | measured | 0.0000 | 0.1100 | Valid | 0.215 | $0.000017 | |
| policy-060 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.239 | $0.000021 | |
| policy-060 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 0.987 | $0.000046 | |
| policy-051 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.240 | $0.000045 | |
| policy-051 | policy | Jev | measured | 0.0000 | 0.1100 | Valid | 0.318 | $0.000020 | |
| probability-060 | probability | Luna | measured | 0.4957 | 0.4467 | Valid | 1.581 | $0.000044 | |
| probability-060 | probability | Jev | measured | 0.4957 | 0.5300 | Valid | 0.217 | $0.000020 | |
| probability-015 | probability | Luna | measured | 0.2452 | 0.2426 | Valid | 1.121 | $0.000040 | |
| probability-015 | probability | Jev | measured | 0.2452 | 0.5900 | Valid | 0.288 | $0.000018 | |
| policy-052 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.265 | $0.000021 | |
| policy-052 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 0.941 | $0.000046 | |
| policy-019 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.008 | $0.000046 | |
| policy-019 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.246 | $0.000021 | |
| boolq-dev-2217 | boolq | Jev | measured | 0.0000 | 0.0400 | Valid | 0.223 | $0.000019 | |
| boolq-dev-2217 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 0.950 | $0.000042 | |
| boolq-dev-2011 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 0.972 | $0.000054 | |
| boolq-dev-2011 | boolq | Jev | measured | 1.0000 | 0.9900 | Valid | 0.223 | $0.000024 | |
| probability-010 | probability | Jev | measured | 0.5000 | 0.5100 | Valid | 0.227 | $0.000018 | |
| probability-010 | probability | Luna | measured | 0.5000 | 0.5000 | Valid | 1.407 | $0.000040 | |
| boolq-dev-0877 | boolq | Jev | measured | 1.0000 | 0.9300 | Valid | 0.274 | $0.000021 | |
| boolq-dev-0877 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.018 | $0.000046 | |
| policy-032 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.249 | $0.000021 | |
| policy-032 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.113 | $0.000046 | |
| policy-017 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.268 | $0.000020 | |
| policy-017 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 2.314 | $0.000045 | |
| policy-008 | policy | Jev | measured | 0.0000 | 0.0600 | Valid | 0.261 | $0.000021 | |
| policy-008 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.006 | $0.000046 | |
| boolq-dev-0065 | boolq | Jev | measured | 1.0000 | 0.9600 | Valid | 0.381 | $0.000017 | |
| boolq-dev-0065 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.104 | $0.000037 | |
| probability-045 | probability | Luna | measured | 0.5000 | 0.5000 | Valid | 0.998 | $0.000040 | |
| probability-045 | probability | Jev | measured | 0.5000 | 0.5100 | Valid | 0.252 | $0.000018 | |
| probability-046 | probability | Luna | measured | 0.1413 | 0.0851 | Valid | 1.258 | $0.000044 | |
| probability-046 | probability | Jev | measured | 0.1413 | 0.1500 | Valid | 0.248 | $0.000020 | |
| probability-001 | probability | Jev | measured | 0.5000 | 0.5300 | Valid | 0.283 | $0.000018 | |
| probability-001 | probability | Luna | measured | 0.5000 | 0.5000 | Valid | 1.274 | $0.000040 | |
| probability-051 | probability | Jev | measured | 0.5000 | 0.5200 | Valid | 0.237 | $0.000018 | |
| probability-051 | probability | Luna | measured | 0.5000 | 0.5000 | Valid | 1.156 | $0.000040 | |
| policy-011 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.238 | $0.000021 | |
| policy-011 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.729 | $0.000046 | |
| policy-010 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.013 | $0.000045 | |
| policy-010 | policy | Jev | measured | 0.0000 | 0.0700 | Valid | 0.319 | $0.000020 | |
| probability-043 | probability | Jev | measured | 0.9259 | 0.8600 | Valid | 0.220 | $0.000020 | |
| probability-043 | probability | Luna | measured | 0.9259 | 0.9200 | Valid | 1.415 | $0.000044 | |
| policy-018 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.254 | $0.000045 | |
| policy-018 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.263 | $0.000020 | |
| policy-009 | policy | Jev | measured | 0.0000 | 0.0900 | Valid | 0.228 | $0.000020 | |
| policy-009 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.259 | $0.000045 | |
| boolq-dev-0659 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 0.874 | $0.000037 | |
| boolq-dev-0659 | boolq | Jev | measured | 1.0000 | 0.9900 | Valid | 0.231 | $0.000017 | |
| probability-034 | probability | Luna | measured | 0.9800 | 0.9800 | Valid | 1.044 | $0.000040 | |
| probability-034 | probability | Jev | measured | 0.9800 | 0.9600 | Valid | 0.233 | $0.000018 | |
| boolq-dev-2951 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.100 | $0.000058 | |
| boolq-dev-2951 | boolq | Jev | measured | 1.0000 | 0.9700 | Valid | 0.259 | $0.000026 | |
| probability-053 | probability | Jev | measured | 0.4991 | 0.6000 | Valid | 0.268 | $0.000018 | |
| probability-053 | probability | Luna | measured | 0.4991 | 0.5037 | Valid | 1.316 | $0.000039 | |
| probability-059 | probability | Jev | measured | 0.2251 | 0.2000 | Valid | 0.311 | $0.000020 | |
| probability-059 | probability | Luna | measured | 0.2251 | 0.2489 | Valid | 1.389 | $0.000044 | |
| boolq-dev-1739 | boolq | Jev | measured | 1.0000 | 0.8900 | Valid | 0.286 | $0.000020 | |
| boolq-dev-1739 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.409 | $0.000044 | |
| policy-058 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.213 | $0.000046 | |
| policy-058 | policy | Jev | measured | 0.0000 | 0.0800 | Valid | 0.221 | $0.000021 | |
| policy-053 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.331 | $0.000020 | |
| policy-053 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.156 | $0.000045 | |
| probability-026 | probability | Jev | measured | 0.8750 | 0.8900 | Valid | 0.219 | $0.000018 | |
| probability-026 | probability | Luna | measured | 0.8750 | 0.8750 | Valid | 1.007 | $0.000039 | |
| probability-009 | probability | Jev | measured | 1.0000 | 0.9900 | Valid | 0.270 | $0.000018 | |
| probability-009 | probability | Luna | measured | 1.0000 | 1.0000 | Valid | 1.004 | $0.000039 | |
| boolq-dev-0308 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.181 | $0.000043 | |
| boolq-dev-0308 | boolq | Jev | measured | 0.0000 | 0.0500 | Valid | 0.255 | $0.000019 | |
| probability-050 | probability | Luna | measured | 0.5167 | 0.5167 | Valid | 1.158 | $0.000042 | |
| probability-050 | probability | Jev | measured | 0.5167 | 0.5500 | Valid | 0.349 | $0.000018 | |
| boolq-dev-1296 | boolq | Jev | measured | 0.0000 | 0.3900 | Valid | 0.372 | $0.000018 | |
| boolq-dev-1296 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.168 | $0.000039 | |
| policy-044 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.214 | $0.000021 | |
| policy-044 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 0.986 | $0.000046 | |
| policy-035 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 0.928 | $0.000045 | |
| policy-035 | policy | Jev | measured | 0.0000 | 0.0600 | Valid | 0.235 | $0.000020 | |
| policy-031 | policy | Jev | measured | 1.0000 | 0.9900 | Valid | 0.329 | $0.000020 | |
| policy-031 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.263 | $0.000045 | |
| probability-035 | probability | Jev | measured | 0.2000 | 0.2200 | Valid | 0.291 | $0.000018 | |
| probability-035 | probability | Luna | measured | 0.2000 | 0.2000 | Valid | 1.281 | $0.000040 | |
| policy-026 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.242 | $0.000021 | |
| policy-026 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.166 | $0.000046 | |
| probability-049 | probability | Luna | measured | 0.5772 | 0.5467 | Valid | 1.008 | $0.000044 | |
| probability-049 | probability | Jev | measured | 0.5772 | 0.6300 | Valid | 0.238 | $0.000020 | |
| boolq-dev-2426 | boolq | Jev | measured | 1.0000 | 0.9800 | Valid | 0.236 | $0.000022 | |
| boolq-dev-2426 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 0.958 | $0.000050 | |
| probability-041 | probability | Jev | measured | 0.0000 | 0.0300 | Valid | 0.257 | $0.000018 | |
| probability-041 | probability | Luna | measured | 0.0000 | 0.0000 | Valid | 1.106 | $0.000038 | |
| policy-045 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.190 | $0.000046 | |
| policy-045 | policy | Jev | measured | 0.0000 | 0.0700 | Valid | 0.232 | $0.000021 | |
| policy-006 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.265 | $0.000046 | |
| policy-006 | policy | Jev | measured | 0.0000 | 0.1100 | Valid | 0.327 | $0.000021 | |
| boolq-dev-3115 | boolq | Luna | measured | 1.0000 | 0.9800 | Valid | 1.143 | $0.000047 | |
| boolq-dev-3115 | boolq | Jev | measured | 1.0000 | 0.9200 | Valid | 0.236 | $0.000021 | |
| policy-038 | policy | Jev | measured | 0.0000 | 0.0300 | Valid | 0.246 | $0.000021 | |
| policy-038 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.389 | $0.000046 | |
| boolq-dev-1662 | boolq | Jev | measured | 1.0000 | 0.9700 | Valid | 0.259 | $0.000021 | |
| boolq-dev-1662 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.042 | $0.000046 | |
| policy-003 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.257 | $0.000046 | |
| policy-003 | policy | Jev | measured | 0.0000 | 0.0500 | Valid | 0.227 | $0.000021 | |
| boolq-dev-2597 | boolq | Luna | measured | 1.0000 | 0.2500 | Valid | 0.970 | $0.000040 | |
| boolq-dev-2597 | boolq | Jev | measured | 1.0000 | 0.9600 | Valid | 0.224 | $0.000018 | |
| probability-002 | probability | Jev | measured | 0.7863 | 0.7700 | Valid | 0.235 | $0.000017 | |
| probability-002 | probability | Luna | measured | 0.7863 | 0.8194 | Valid | 1.234 | $0.000039 | |
| policy-059 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 0.900 | $0.000046 | |
| policy-059 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.392 | $0.000021 | |
| probability-029 | probability | Jev | measured | 0.8894 | 0.8700 | Valid | 0.232 | $0.000020 | |
| probability-029 | probability | Luna | measured | 0.8894 | 0.9250 | Valid | 1.338 | $0.000044 | |
| policy-050 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.204 | $0.000046 | |
| policy-050 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.307 | $0.000021 | |
| boolq-dev-1381 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.166 | $0.000037 | |
| boolq-dev-1381 | boolq | Jev | measured | 0.0000 | 0.1500 | Valid | 0.317 | $0.000017 | |
| policy-048 | policy | Jev | measured | 0.0000 | 0.1300 | Valid | 0.265 | $0.000021 | |
| policy-048 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.267 | $0.000046 | |
| boolq-dev-0210 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.230 | $0.000041 | |
| boolq-dev-0210 | boolq | Jev | measured | 1.0000 | 0.9600 | Valid | 0.352 | $0.000019 | |
| boolq-dev-2849 | boolq | Jev | measured | 0.0000 | 0.4600 | Valid | 0.226 | $0.000018 | |
| boolq-dev-2849 | boolq | Luna | measured | 0.0000 | 0.0500 | Valid | 1.268 | $0.000042 | |
| probability-006 | probability | Luna | measured | 0.2800 | 0.5600 | Valid | 0.923 | $0.000039 | |
| probability-006 | probability | Jev | measured | 0.2800 | 0.4100 | Valid | 0.272 | $0.000018 | |
| boolq-dev-2774 | boolq | Luna | measured | 0.0000 | 0.0500 | Valid | 1.185 | $0.000050 | |
| boolq-dev-2774 | boolq | Jev | measured | 0.0000 | 0.3900 | Valid | 0.222 | $0.000023 | |
| probability-004 | probability | Jev | measured | 0.2000 | 0.2200 | Valid | 0.257 | $0.000018 | |
| probability-004 | probability | Luna | measured | 0.2000 | 0.2000 | Valid | 1.562 | $0.000040 | |
| boolq-dev-0852 | boolq | Luna | measured | 0.0000 | 0.5000 | Valid | 0.978 | $0.000033 | |
| boolq-dev-0852 | boolq | Jev | measured | 0.0000 | 0.1100 | Valid | 0.576 | $0.000014 | |
| boolq-dev-2868 | boolq | Jev | measured | 0.0000 | 0.0700 | Valid | 0.226 | $0.000030 | |
| boolq-dev-2868 | boolq | Luna | measured | 0.0000 | 1.0000 | Valid | 1.192 | $0.000067 | |
| boolq-dev-3247 | boolq | Jev | measured | 1.0000 | 0.9900 | Valid | 0.234 | $0.000018 | |
| boolq-dev-3247 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.154 | $0.000039 | |
| boolq-dev-1603 | boolq | Jev | measured | 0.0000 | 0.0400 | Valid | 0.267 | $0.000020 | |
| boolq-dev-1603 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.047 | $0.000045 | |
| probability-058 | probability | Jev | measured | 0.6691 | 0.6900 | Valid | 0.241 | $0.000020 | |
| probability-058 | probability | Luna | measured | 0.6691 | 0.7989 | Valid | 1.143 | $0.000044 | |
| policy-014 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.926 | $0.000045 | |
| policy-014 | policy | Jev | measured | 0.0000 | 0.0900 | Valid | 0.244 | $0.000020 | |
| policy-020 | policy | Jev | measured | 0.0000 | 0.0900 | Valid | 0.234 | $0.000021 | |
| policy-020 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.210 | $0.000046 | |
| boolq-dev-2116 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.306 | $0.000043 | |
| boolq-dev-2116 | boolq | Jev | measured | 1.0000 | 0.9900 | Valid | 0.221 | $0.000019 | |
| policy-033 | policy | Jev | measured | 0.0000 | 0.1300 | Valid | 0.244 | $0.000021 | |
| policy-033 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.214 | $0.000046 | |
| policy-029 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 0.898 | $0.000046 | |
| policy-029 | policy | Jev | measured | 0.0000 | 0.0900 | Valid | 0.220 | $0.000021 | |
| policy-055 | policy | Jev | measured | 1.0000 | 0.9700 | Valid | 0.221 | $0.000021 | |
| policy-055 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.244 | $0.000046 | |
| policy-005 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.234 | $0.000020 | |
| policy-005 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.199 | $0.000045 | |
| policy-030 | policy | Jev | measured | 0.0000 | 0.0700 | Valid | 0.238 | $0.000021 | |
| policy-030 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.048 | $0.000046 | |
| probability-012 | probability | Jev | measured | 0.8890 | 0.8600 | Valid | 0.224 | $0.000018 | |
| probability-012 | probability | Luna | measured | 0.8890 | 0.8654 | Valid | 1.333 | $0.000040 | |
| policy-022 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.073 | $0.000046 | |
| policy-022 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.219 | $0.000021 | |
| probability-013 | probability | Luna | measured | 0.9500 | 0.9500 | Valid | 1.142 | $0.000040 | |
| probability-013 | probability | Jev | measured | 0.9500 | 0.9400 | Valid | 0.234 | $0.000018 | |
| boolq-dev-3134 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.563 | $0.000036 | |
| boolq-dev-3134 | boolq | Jev | measured | 1.0000 | 0.9800 | Valid | 0.228 | $0.000016 | |
| boolq-dev-1953 | boolq | Luna | measured | 0.0000 | 0.0000 | Valid | 1.470 | $0.000038 | |
| boolq-dev-1953 | boolq | Jev | measured | 0.0000 | 0.0200 | Valid | 0.279 | $0.000017 | |
| policy-054 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.027 | $0.000046 | |
| policy-054 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.229 | $0.000021 | |
| boolq-dev-2818 | boolq | Jev | measured | 1.0000 | 0.8800 | Valid | 0.264 | $0.000019 | |
| boolq-dev-2818 | boolq | Luna | measured | 1.0000 | 0.5000 | Valid | 1.021 | $0.000044 | |
| policy-004 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.642 | $0.000045 | |
| policy-004 | policy | Jev | measured | 0.0000 | 0.0700 | Valid | 0.230 | $0.000020 | |
| boolq-dev-0383 | boolq | Jev | measured | 1.0000 | 0.9900 | Valid | 0.224 | $0.000017 | |
| boolq-dev-0383 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 1.119 | $0.000037 | |
| probability-037 | probability | Luna | measured | 0.0333 | 0.0333 | Valid | 1.499 | $0.000043 | |
| probability-037 | probability | Jev | measured | 0.0333 | 0.0600 | Valid | 0.234 | $0.000018 | |
| probability-036 | probability | Jev | measured | 0.1048 | 0.2800 | Valid | 0.283 | $0.000017 | |
| probability-036 | probability | Luna | measured | 0.1048 | 0.1000 | Valid | 1.063 | $0.000038 | |
| probability-003 | probability | Luna | measured | 0.0000 | 0.0000 | Valid | 1.301 | $0.000039 | |
| probability-003 | probability | Jev | measured | 0.0000 | 0.0100 | Valid | 0.254 | $0.000018 | |
| boolq-dev-2665 | boolq | Luna | measured | 0.0000 | 0.9700 | Valid | 0.960 | $0.000046 | |
| boolq-dev-2665 | boolq | Jev | measured | 0.0000 | 0.2200 | Valid | 0.338 | $0.000020 | |
| boolq-dev-1842 | boolq | Luna | measured | 1.0000 | 0.0000 | Valid | 1.265 | $0.000041 | |
| boolq-dev-1842 | boolq | Jev | measured | 1.0000 | 0.0300 | Valid | 0.352 | $0.000018 | |
| probability-008 | probability | Luna | measured | 0.9692 | 0.9677 | Valid | 1.448 | $0.000040 | |
| probability-008 | probability | Jev | measured | 0.9692 | 0.9600 | Valid | 0.220 | $0.000018 | |
| probability-047 | probability | Luna | measured | 0.7708 | 0.8000 | Valid | 1.075 | $0.000044 | |
| probability-047 | probability | Jev | measured | 0.7708 | 0.7300 | Valid | 0.222 | $0.000020 | |
| policy-024 | policy | Jev | measured | 0.0000 | 0.0900 | Valid | 0.219 | $0.000020 | |
| policy-024 | policy | Luna | measured | 0.0000 | 0.0000 | Valid | 1.049 | $0.000045 | |
| boolq-dev-0468 | boolq | Jev | measured | 1.0000 | 0.9900 | Valid | 0.225 | $0.000018 | |
| boolq-dev-0468 | boolq | Luna | measured | 1.0000 | 0.9900 | Valid | 1.136 | $0.000042 | |
| boolq-dev-0587 | boolq | Jev | measured | 1.0000 | 0.1500 | Valid | 0.234 | $0.000019 | |
| boolq-dev-0587 | boolq | Luna | measured | 1.0000 | 0.0000 | Valid | 1.318 | $0.000041 | |
| boolq-dev-2630 | boolq | Luna | measured | 1.0000 | 1.0000 | Valid | 0.992 | $0.000039 | |
| boolq-dev-2630 | boolq | Jev | measured | 1.0000 | 0.9800 | Valid | 0.240 | $0.000018 | |
| probability-019 | probability | Jev | measured | 0.9333 | 0.9100 | Valid | 0.267 | $0.000018 | |
| probability-019 | probability | Luna | measured | 0.9333 | 0.9333 | Valid | 2.471 | $0.000041 | |
| policy-037 | policy | Jev | measured | 1.0000 | 0.9800 | Valid | 0.270 | $0.000021 | |
| policy-037 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.001 | $0.000046 | |
| probability-054 | probability | Jev | measured | 0.0238 | 0.1500 | Valid | 0.246 | $0.000017 | |
| probability-054 | probability | Luna | measured | 0.0238 | 0.0238 | Valid | 1.023 | $0.000040 | |
| policy-042 | policy | Luna | measured | 1.0000 | 1.0000 | Valid | 1.174 | $0.000046 | |
| policy-042 | policy | Jev | measured | 1.0000 | 0.6000 | Valid | 0.244 | $0.000021 |
Metric definitions and statistical method
- success
- valid=true with a finite probability and valid target in [0,1]; label targets must be 0 or 1
- accuracy
- Fraction correct among valid responses, predicting yes when p_yes >= 0.5
- accuracy_all_requests
- Correct valid responses divided by every measured request; invalid requests count as failures
- brier
- Mean (p_yes - observed_label)^2; lower is better
- log_loss
- Mean binary negative log likelihood in natural-log units; p clipped to [1e-15, 1-1e-15]
- mae
- Mean absolute difference between reported p_yes and exact reference probability
- rmse
- Square root of mean squared error against exact reference probabilities
- excess_brier
- Mean (p_yes - true_probability)^2: expected Brier above the optimal probability forecast
- expected_brier
- Mean [(p_yes - true_probability)^2 + true_probability*(1-true_probability)]
- latency
- End-to-end request time through receipt of a complete response; main table includes valid responses only
- cost
- Recorded billed USD across all requests in the stated phase, including failed/invalid requests when billed
- cost_per_1000
- Measured total billed cost / measured request count * 1000; withheld if any bill is missing
- cost_per_correct
- All measured billed cost / number of correct valid label responses; withheld for missing bills or zero correct
- bootstrap
- Percentile 95% paired bootstrap by case identity; quality repeats averaged within case; latency repeats use median; no CI for fewer than 2 pairs
Recorded run metadata
{
"models": {
"Jev": "typesafe/jev-1.13",
"Luna": "openai/gpt-6-luna"
},
"created_at": "2026-10-02T08:42:23.244948+00:00",
"seed": 20261002,
"cases": 180,
"calls_planned": 360,
"budget_usd": 0.999912658,
"luna_reasoning": "none",
"luna_provider": "OpenAI",
"protocol": "One non-streaming request per model per case; sequential paired random order; one shared persistent HTTPS connection; no automatic retries; two warmups per model excluded from measured results.",
"cases_sha256": "60e6db60c72869936548c3c4620aa3b070cf12e516f3be42a61a306d0f592660",
"runner_sha256": "3a05cb20816fbc5b7c30a065429a7365afa9583f273d683bf69784f4af3e9ad7",
"status": "complete",
"charged_usd": 0.011550069999999994,
"calls_completed": 364,
"finished_at": "2026-10-02T08:46:47.575786+00:00"
}