🚧 Work in progress — this arena is under active
development. Leaderboards, fighters and data are incomplete and may change
or reset without notice; don't rely on these numbers yet.
⚔️ Blind Battle
Two masked fighters interpret the same input. Pick the better output —
your vote opens a GitHub issue, no arena account needed.
Loading battles pool…
Try another league or language. No battles available — run the assemble.yml GitHub Action to assemble battles from prediction datasets.
Languages you speak:Modalities:
heard utterance
Which interpretation should a voice assistant act on? “No match” means it would ask the user to rephrase.
Candidate A
VS
Candidate B
Voting reveals the fighters and opens a prefilled GitHub issue — submit it to record your vote.