Qwen3.8 27B (Chutes TEE)
Data: JevBench v1.6.1, published 2026-10-06
Capability 98.2 · API offering, ranked on the API leaderboard · outside the Jev-class caps (cost 67.4× Jev (v1.5 reference); latency 11.5× 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.
open weights
intelligence
96.4
calibration
100.0
speed
57.4
cost
0.0
Composite (secondary)
0.0 · Composite rank #27 of 27 on the API leaderboard
Cost and measurement conditions
$2.178 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: 7.1 s
none (API)
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 | 97.7 |
| noul | 96.5 | 95.2 |
| score | 88.8 | 87.7 |
Capability by subject topic
| Rules, policy & law | 98.3 | n=1124 |
|---|---|---|
| Coding & software | 90.8 | n=201 |
| Math & numbers | 91.6 | n=268 |
| Finance & commerce | 98.9 | n=177 |
| Support & operations | 99.7 | n=108 |
| Everyday language | 96.7 | n=130 |
| Safety & security | 99.6 | n=83 |
Use cases (TypeSafe categories)
| Model routing | 92.0 | n=267 |
|---|---|---|
| Legal & compliance | 96.2 | n=266 |
| Customer support | 99.9 | n=155 |
| E-commerce | 99.9 | n=87 |
| Insurance claims | 100.0 | n=91 |
| Risk assessment | 100.0 | n=87 |
| Financial crime | 100.0 | n=75 |
| Feature extraction | 94.7 | n=89 |
| Lead generation | 98.1 | n=86 |
| Recruiting | 96.5 | n=87 |
| Knowledge graphs | 98.6 | n=81 |
| LLM guardrails | 98.6 | n=82 |
| Moderation | 99.8 | n=68 |
| Code linting | 97.9 | n=80 |
| Search & retrieval | 97.3 | n=83 |
| Science | 96.1 | n=82 |
| Advertising | 99.9 | n=85 |
| Gaming | 99.7 | n=79 |
| Demand forecasting | 98.5 | 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 96.8 · composite 0.0 · cost $2.178 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.