GLiNER2.5-Decide vs Kev 4B r10
Kev 4B r10 is more accurate on the Jev Decision Index, Kev 4B r10 is better calibrated. Choose on whichever of accuracy, calibration and cost matters most for your decisions.
| Spec | GLiNER2.5-Decide | Kev 4B r10 |
|---|---|---|
| Provider | Fastino | Jared Palmer |
| Decision Index score | 10.1 | 34.9 |
| Calibration error (ECE) | 0.088 | 0.032 |
| Median latency (p50) | 23 ms | 59 ms |
| Cost per 1,000 decisions | — | $0.0084 |
| Parameters | 486M | 5B |
| Open weights | Yes | Yes |
| Question types | Choice (one of N options), Score (ordered rubric), Yes / no, Extraction (typed spans) | Choice (one of N options), Score (ordered rubric), Yes / no |
| Mechanism | Span encoder (GLiNER family) | Decoder LM with a decision readout |
| License | apache-2.0 | apache-2.0 |
FAQ
Is GLiNER2.5-Decide or Kev 4B r10 more accurate?
Kev 4B r10 scores higher on the Jev Decision Index (34.9 vs 10.1).
Which is better calibrated, GLiNER2.5-Decide or Kev 4B r10?
Kev 4B r10 has the lower expected calibration error (0.032 vs 0.088), so its stated confidence is closer to how often it is right.
Is GLiNER2.5-Decide or Kev 4B r10 cheaper per 1,000 decisions?
A hosted price is not recorded for both models; open-weight models cost the compute you run them on.
Cost per 1,000 decisions assumes 200 input tokens per decision. Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.