SimpleJev Qwen3.8-27B
Data: JevBench v1.6.1, published 2026-10-06
Capability 71.6 · API offering, ranked on the API leaderboard · outside the Jev-class caps (cost 21.3× 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.
Apache-2.0 (Qwen weights); repository licence not stated
intelligence
52.9
calibration
90.3
speed
76.9
cost
14.9
Composite (secondary)
3.2 · Composite rank #23 of 27 on the API leaderboard
Cost and measurement conditions
$0.687 per 1,000 decisions (estimated)
v1.5.4 published cost per 1,000 decisions carried (pricing rules unchanged; v1.6 item lengths differ): ESTIMATE: exact base-model market reference (operator tariff excluded by 30-day rule)
p50 latency: 1.2 s
x2 (demo)
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 | 77.1 | 61.5 |
| noul | 39.8 | 31.4 |
| score | 57.2 | 34.8 |
Capability by subject topic
| Rules, policy & law | 40.7 | n=1133 |
|---|---|---|
| Coding & software | 73.3 | n=202 |
| Math & numbers | 2.2 | n=269 |
| Finance & commerce | 36.7 | n=178 |
| Support & operations | 45.4 | n=109 |
| Everyday language | 83.2 | n=130 |
| Safety & security | 67.9 | n=84 |
Use cases (TypeSafe categories)
| Model routing | 73.7 | n=267 |
|---|---|---|
| Legal & compliance | 51.8 | n=269 |
| Customer support | 40.2 | n=156 |
| E-commerce | 35.6 | n=88 |
| Insurance claims | 22.3 | n=92 |
| Risk assessment | 31.3 | n=88 |
| Financial crime | 15.9 | n=76 |
| Feature extraction | 22.0 | n=90 |
| Lead generation | 48.4 | n=86 |
| Recruiting | 0.0 | n=88 |
| Knowledge graphs | 38.0 | n=82 |
| LLM guardrails | 82.9 | n=82 |
| Moderation | 66.4 | n=69 |
| Code linting | 59.5 | n=81 |
| Search & retrieval | 91.0 | n=84 |
| Science | 2.5 | n=82 |
| Advertising | 31.0 | n=85 |
| Gaming | 21.9 | n=79 |
| Demand forecasting | 0.0 | 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 80.0 · composite 3.4 · cost $0.687 per 1,000 decisions (estimated)
Last measured 2026-09-28, measurement release v1.5.0. Older method and scale; the current figures above replace it.