Small decisions. Interesting possibilities.Submit contentSubmit
OpenRouter’s Ori Eval preview

OpenRouter’s Ori Eval

1/ Jev, a decision model by @typesafeai, sparked a burst of projects and discussion. We tested it using Ori Eval against popular LLMs on OpenRouter at judging. Jev was >5x faster than the next fastest model, and even its slowest requests beat every other model's median.

Added to Jevfast

How it uses Jev

Input: model-judging requests in the Ori Eval benchmark. Jev is evaluated as a judge against other models; the post reports speed comparisons but does not say how its judgment is scored or consumed. Exact Jev request and answer shape are undisclosed.

Why it is interesting

The example treats judging as a measurable model task. The post provides relative speed claims, but not the judge prompt, test set, or quality results.