wity-1 (Wity, reasoning off)v1.7 · 10 s rule
Data: JevBench v1.8.18, published 2026-11-20 · method v1.7 (10 s rule), re-scored from the recorded per-decision times
Capability 63.0 · API offering, listed but not ranked on the API leaderboard
registry ranked=false or no composite
Last measured: 2026-10-06 · measurement release: v1.6.1
proprietary hosted API
intelligence
44.1
calibration
81.8
speed
87.6
cost
58.8
Composite (secondary)
49.2 · Not ranked on this board
Cost and measurement conditions
$0.024 per 1,000 decisions (public list price)
Wity's stated API tariff USD 0.042 per 1M input tokens, output free; measured on the v1.6.1 common cost basis (all items except those outside the Jev reference input range) from this run's token usage
p50 latency: 0.3 s
none (API)
operator's hosted production API, measured as is; API default is off, our request sets reasoning=off; labelled configuration variant. The first approximately 70 decisions ran alongside two other streams; the remainder ran alone.
Measured latency over 1,500 decisions: p50 0.345 s · p95 0.502 s
0 decisions over 10 s.
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.
Breakdowns (not part of the score)
Chance-corrected competence (0 = chance, 100 = perfect) by category on S+P+L1+L2+L3. Raw values, not equated; the same cells feed the compare view.
Capability by subject topic
| Rules, policy & law | 33.4 | n=1616 |
|---|---|---|
| Coding & software | 57.1 | n=382 |
| Math & numbers | 11.4 | n=370 |
| Finance & commerce | 30.7 | n=223 |
| Support & operations | 36.3 | n=160 |
| Everyday language | 60.5 | n=145 |
| Safety & security | 58.6 | n=109 |
Use cases (TypeSafe categories)
| Model routing | 59.7 | n=524 |
|---|---|---|
| Legal & compliance | 50.1 | n=515 |
| Customer support | 37.8 | n=240 |
| E-commerce | 24.4 | n=119 |
| Insurance claims | 19.5 | n=119 |
| Risk assessment | 47.6 | n=118 |
| Financial crime | 32.0 | n=101 |
| Feature extraction | 14.5 | n=99 |
| Lead generation | 40.4 | n=95 |
| Recruiting | 8.0 | n=94 |
| Knowledge graphs | 3.7 | n=94 |
| LLM guardrails | 32.5 | n=93 |
| Moderation | 46.9 | n=93 |
| Code linting | 23.4 | n=90 |
| Search & retrieval | 57.8 | n=90 |
| Science | 3.6 | n=90 |
| Advertising | 24.9 | n=88 |
| Gaming | 9.8 | n=86 |
| Demand forecasting | 9.3 | n=83 |
Cells need at least 30 answered items; † fewer than 30 (indicative only; drawn as hollow, lighter points on the compare radars).
JevBench v1.6.43 values (method v1.6, before the 10 s rule)
Capability 62.97 · Composite 49.37
intelligence 44.11 · calibration 81.83 · speed 89.56 · cost 58.83
v1.6 speed basis (124 items): p50 0.282 s · p95 0.393 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.