d1 (Liquid AI)
Data: JevBench v1.6.39, published 2026-10-22
Capability 74.4 · API offering, ranked on the API leaderboard · Capability rank #3 of 14 within the Jev-class caps
Measured 2026-10-06 on the full v1.6.1 set (1500 items), scored like the other full-set API rows, not equated.
proprietary hosted API
intelligence
62.1
calibration
86.8
speed
89.2
cost
62.8
Composite (secondary)
73.0 · Composite rank #2 of 27 on the API leaderboard
Cost and measurement conditions
$0.017 per 1,000 decisions (public tariff)
Liquid AI d1 public API tariff USD 0.04 per 1M input tokens, no output tokens (Vercel AI Gateway listing https://vercel.com/ai-gateway/models/d1, read 6 Oct 2026; equals Liquid API usage.cost); measured on the v1.6.1 common cost basis from this run's token usage
p50 latency: 0.3 s
none (API)
Breakdowns (not part of the score)
Chance-corrected competence (0 = chance, 100 = perfect) on the full set (1500 items). Raw values, not equated. Category cells use S+P+L1+L2+L3; per-type/tier cells use the original basis above. Radars against Jev.
| Request type | Open | Sealed |
|---|---|---|
| choice | 75.5 | 71.2 |
| noul | 61.4 | 46.4 |
| score | 57.7 | 60.2 |
Capability by subject topic
| Rules, policy & law | 53.5 | n=1616 |
|---|---|---|
| Coding & software | 64.2 | n=382 |
| Math & numbers | 11.2 | n=370 |
| Finance & commerce | 42.9 | n=223 |
| Support & operations | 52.2 | n=160 |
| Everyday language | 77.2 | n=145 |
| Safety & security | 57.0 | n=109 |
Use cases (TypeSafe categories)
| Model routing | 67.5 | n=524 |
|---|---|---|
| Legal & compliance | 71.2 | n=515 |
| Customer support | 56.5 | n=240 |
| E-commerce | 50.3 | n=119 |
| Insurance claims | 37.9 | n=119 |
| Risk assessment | 39.6 | n=118 |
| Financial crime | 26.9 | n=101 |
| Feature extraction | 28.0 | n=99 |
| Lead generation | 58.4 | n=95 |
| Recruiting | 18.6 | n=94 |
| Knowledge graphs | 39.3 | n=94 |
| LLM guardrails | 74.1 | n=93 |
| Moderation | 61.5 | n=93 |
| Code linting | 39.3 | n=90 |
| Search & retrieval | 78.9 | n=90 |
| Science | 30.9 | n=90 |
| Advertising | 38.2 | n=88 |
| Gaming | 0.0 | n=86 |
| Demand forecasting | 0.0 | n=83 |
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 74.4 · composite 73.0 · cost $0.017 per 1,000 decisions (not published)
Last measured 2026-10-06, measurement release v1.6.1. Older method and scale; the current figures above replace it.