Nace.AI Drex v2.0 (35B-A3B) vs Celeris-1 Decision
Celeris-1 Decision is cheaper per 1,000 decisions. Choose on whichever of accuracy, calibration and cost matters most for your decisions.
| Spec | Nace.AI Drex v2.0 (35B-A3B) | Celeris-1 Decision |
|---|---|---|
| Provider | Nace.AI | Celeris |
| Decision Index score | 55.9 | — |
| Calibration error (ECE) | 0.018 | — |
| Median latency (p50) | 276 ms | — |
| Cost per 1,000 decisions | $0.012 | $0.0080 |
| Parameters | 35B | — |
| Open weights | Yes | No |
| 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 | Diffusion LM |
| License | nace-ai-open-rail-m | proprietary |
FAQ
Is Nace.AI Drex v2.0 (35B-A3B) or Celeris-1 Decision more accurate?
At least one of the two has no Jev Decision Index score yet, so their accuracy cannot be compared here.
Which is better calibrated, Nace.AI Drex v2.0 (35B-A3B) or Celeris-1 Decision?
No calibration difference between Nace.AI Drex v2.0 (35B-A3B) and Celeris-1 Decision is recorded yet.
Is Nace.AI Drex v2.0 (35B-A3B) or Celeris-1 Decision cheaper per 1,000 decisions?
Celeris-1 Decision is cheaper: about $0.0080 vs $0.012 per 1,000 decisions at 200 input tokens each.
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.