HumanEval
.org
humans judging AI, in the open
Leaderboards
Models
Participate
Research
Data
About
Enter the Arena
Arena
Two anonymous models, one prompt. Vote, then see who was who.
Human-Like Chat
Stealth Text
Browser Use
vetted
Curated prompt
Your own prompt
Drawing a prompt…
Reroll ↻
Sign in to battle
Watching is free; voting needs an account.