GLiNER2.5-Decide vs Matilda Jev v1.5
Matilda Jev v1.5 is more accurate on the Jev Decision Index, Matilda Jev v1.5 is better calibrated. Choose on whichever of accuracy, calibration and cost matters most for your decisions.
| Spec | GLiNER2.5-Decide | Matilda Jev v1.5 |
|---|---|---|
| Provider | Fastino | Maincode |
| Decision Index score | 10.1 | 59.2 |
| Calibration error (ECE) | 0.088 | 0.077 |
| Median latency (p50) | 23 ms | 89 ms |
| Cost per 1,000 decisions | — | — |
| Parameters | 486M | 26B |
| 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 Matilda Jev v1.5 more accurate?
Matilda Jev v1.5 scores higher on the Jev Decision Index (59.2 vs 10.1).
Which is better calibrated, GLiNER2.5-Decide or Matilda Jev v1.5?
Matilda Jev v1.5 has the lower expected calibration error (0.077 vs 0.088), so its stated confidence is closer to how often it is right.
Is GLiNER2.5-Decide or Matilda Jev v1.5 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.