vLLM-SR d3 (27B) vs linne-oak-1-0
vLLM-SR d3 (27B) is more accurate on the Jev Decision Index, vLLM-SR d3 (27B) is better calibrated. Choose on whichever of accuracy, calibration and cost matters most for your decisions.
| Spec | vLLM-SR d3 (27B) | linne-oak-1-0 |
|---|---|---|
| Provider | vLLM Semantic Router | MISA-AICORE |
| Decision Index score | 67.2 | 66.9 |
| Calibration error (ECE) | 0.043 | 0.105 |
| Median latency (p50) | 97 ms | 105 ms |
| Cost per 1,000 decisions | — | — |
| Parameters | 26B | 31B |
| Open weights | Yes | Yes |
| Question types | Choice (one of N options), Score (ordered rubric), Yes / no | Choice (one of N options), Score (ordered rubric), Yes / no |
| Mechanism | Decoder LM with a decision readout | Decoder LM with a decision readout |
| License | apache-2.0 | apache-2.0 |
FAQ
Is vLLM-SR d3 (27B) or linne-oak-1-0 more accurate?
vLLM-SR d3 (27B) scores higher on the Jev Decision Index (67.2 vs 66.9).
Which is better calibrated, vLLM-SR d3 (27B) or linne-oak-1-0?
vLLM-SR d3 (27B) has the lower expected calibration error (0.043 vs 0.105), so its stated confidence is closer to how often it is right.
Is vLLM-SR d3 (27B) or linne-oak-1-0 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.