Explorer
Every vote in the benchmark. Choose a language to read titles and verdicts in it, and open a vote to see the gold answers, the sources and every answer each model gave.
Loading votes
| Date | Vote | Instrument | Government | Outcome | 70B (Featherless) | Public AI "8B" route | Public AI "8B" route, thinking | 70B thinking (Featherless) |
|---|
Run columns show the verdict on the outcome question (F1) in the selected language. For the votes not yet held it is F7, where the right answer is that the vote has not happened.