Submit a model's benchmark run for review, or moderate the queue.
Submit a run Moderation queue
Prefer the command line? Run your evaluation with script.py --publish — if it can't open a browser itself, it prints a code to enter at /device.
script.py --publish
View the live leaderboard →