Darwin-27B-ZTC vs Cloudflare clef-flash
Darwin-27B-ZTC is more accurate on the Jev Decision Index, Cloudflare clef-flash is better calibrated. Choose on whichever of accuracy, calibration and cost matters most for your decisions.
| Spec | Darwin-27B-ZTC | Cloudflare clef-flash |
|---|---|---|
| Provider | FINAL_Bench | Cloudflare |
| Decision Index score | 62.5 | 47.6 |
| Calibration error (ECE) | 0.045 | 0.025 |
| Median latency (p50) | 104 ms | 53 ms |
| Cost per 1,000 decisions | — | — |
| Parameters | 26B | 10B |
| 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 Darwin-27B-ZTC or Cloudflare clef-flash more accurate?
Darwin-27B-ZTC scores higher on the Jev Decision Index (62.5 vs 47.6).
Which is better calibrated, Darwin-27B-ZTC or Cloudflare clef-flash?
Cloudflare clef-flash has the lower expected calibration error (0.025 vs 0.045), so its stated confidence is closer to how often it is right.
Is Darwin-27B-ZTC or Cloudflare clef-flash 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.