GLiNER2.5-Decide vs Hopper 12B (trained)
Hopper 12B (trained) is more accurate on the Jev Decision Index, Hopper 12B (trained) is better calibrated. Choose on whichever of accuracy, calibration and cost matters most for your decisions.
| Spec | GLiNER2.5-Decide | Hopper 12B (trained) |
|---|---|---|
| Provider | Fastino | HopitAI |
| Decision Index score | 10.1 | 45.8 |
| Calibration error (ECE) | 0.088 | 0.076 |
| Median latency (p50) | 23 ms | 290 ms |
| Cost per 1,000 decisions | — | — |
| Parameters | 486M | 12B |
| 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 | research-and-demo |
FAQ
Is GLiNER2.5-Decide or Hopper 12B (trained) more accurate?
Hopper 12B (trained) scores higher on the Jev Decision Index (45.8 vs 10.1).
Which is better calibrated, GLiNER2.5-Decide or Hopper 12B (trained)?
Hopper 12B (trained) has the lower expected calibration error (0.076 vs 0.088), so its stated confidence is closer to how often it is right.
Is GLiNER2.5-Decide or Hopper 12B (trained) 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.