GPT-6 Luna (low reasoning effort)v1.7 · 10 s rule
Data: JevBench v1.8.7, published 2026-11-03 · method v1.7 (10 s rule), re-scored from the recorded per-decision times
Capability 97.1 · API offering, ranked on the API leaderboard · outside the Jev-class caps (cost 3.3× Jev (v1.5 reference); latency 3.1× Jev (v1.5 reference)), below the Capability ranking
Measured 2026-10-06 on A4 ∪ P, 600 items, equated (+2.57 I / +3.35 C); round score-a4-2.
proprietary API
intelligence
95.7
calibration
98.5
speed
71.8
cost
39.0
Composite (secondary)
40.4 · Composite rank #30 of 48 on the API leaderboard
Cost and measurement conditions
$0.108 per 1,000 decisions (public tariff)
measured: token usage of the 6 Oct 2026 A4 u P run x tariff USD 0.1/M input, 0.5/M output (spec gpt-6-luna-low.json), per 1,000 decisions on the v1.6.1 common cost basis (591 of 600 items; the 9 A4 u P items outside the Jev reference's accepted input range are excluded, cost-basis-a4.json)
p50 latency: 1.9 s
none (API)
Measured latency over 600 decisions: p50 1.896 s · p95 3.465 s
3 decisions over 10 s (3 of them answered without error, counted as wrong).
A decision that takes longer than 10 seconds (measured time) counts as a wrong answer and as zero calibration for that item: each type's calibration is multiplied by the share of its decisions answered within 10 s. The limit uses the measured time (raw). The displayed adjusted time for self-hosted rows (2 × + 0.15 s) is an assumption and is not used for the limit.
JevBench v1.6.43 values (method v1.6, before the 10 s rule)
Capability 97.82 · Composite 40.48 · rank #81 on the all-systems board · #28 API
intelligence 96.62 · calibration 99.02 · speed 71.66 · cost 38.98
v1.6 speed basis (124 items): p50 1.854 s · p95 3.683 s
These figures are kept for comparison. The current figures above are re-scored under the 10 s rule from the recorded per-decision times and replace them.
Breakdowns (not part of the score)
Chance-corrected competence (0 = chance, 100 = perfect) on A4 ∪ P (600 items: 300 open + 300 sealed; fewer sealed items than the full set, so wider uncertainty). Raw values, not equated. Category cells use A4+P+L1+L2+L3; per-type/tier cells use the original basis above. Radars against Jev.
| Request type | Open | Sealed |
|---|---|---|
| choice | 97.0 | 100.0 |
| noul | 96.3 | 92.2 |
| score | 85.9 | 87.1 |
Capability by subject topic
| Rules, policy & law | 97.7 | n=1133 |
|---|---|---|
| Coding & software | 89.0 | n=202 |
| Math & numbers | 91.9 | n=269 |
| Finance & commerce | 95.7 | n=178 |
| Support & operations | 99.7 | n=109 |
| Everyday language | 97.5 | n=130 |
| Safety & security | 99.9 | n=84 |
Use cases (TypeSafe categories)
| Model routing | 91.8 | n=267 |
|---|---|---|
| Legal & compliance | 98.0 | n=269 |
| Customer support | 100.0 | n=156 |
| E-commerce | 96.6 | n=88 |
| Insurance claims | 93.2 | n=92 |
| Risk assessment | 95.8 | n=88 |
| Financial crime | 98.2 | n=76 |
| Feature extraction | 97.8 | n=90 |
| Lead generation | 97.6 | n=86 |
| Recruiting | 100.0 | n=88 |
| Knowledge graphs | 97.3 | n=82 |
| LLM guardrails | 94.0 | n=82 |
| Moderation | 97.0 | n=69 |
| Code linting | 95.3 | n=81 |
| Search & retrieval | 99.7 | n=84 |
| Science | 95.7 | n=82 |
| Advertising | 94.9 | n=85 |
| Gaming | 93.8 | n=79 |
| Demand forecasting | 100.0 | n=72 |
Cells need at least 15 answered items; † fewer than 30 (indicative only, listed below the radar in the compare view).
Previous measurement (v1.5.x, not comparable)
Capability 97.1 · composite 40.4 · cost $0.108 per 1,000 decisions (public list price)
Last measured 2026-10-06, measurement release v1.6.1. Older method and scale; the current figures above replace it.