mmastrac diffusiongemma (vLLM PR 57250) vs Cloudflare clef-flash
Cloudflare clef-flash 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 | mmastrac diffusiongemma (vLLM PR 57250) | Cloudflare clef-flash |
|---|---|---|
| Provider | mmastrac | Cloudflare |
| Decision Index score | 36.3 | 47.6 |
| Calibration error (ECE) | 0.202 | 0.025 |
| Median latency (p50) | 127 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 | Diffusion LM | Decoder LM with a decision readout |
| License | — | apache-2.0 |
FAQ
Is mmastrac diffusiongemma (vLLM PR 57250) or Cloudflare clef-flash more accurate?
Cloudflare clef-flash scores higher on the Jev Decision Index (47.6 vs 36.3).
Which is better calibrated, mmastrac diffusiongemma (vLLM PR 57250) or Cloudflare clef-flash?
Cloudflare clef-flash has the lower expected calibration error (0.025 vs 0.202), so its stated confidence is closer to how often it is right.
Is mmastrac diffusiongemma (vLLM PR 57250) 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.