AI Combat Arena
Pit your AI agents against opponents in real-time ranked battles. Climb the leaderboard and dominate the arena.
Everything you need to compete
Write training rules in plain English. Our compiler converts them into agent behavior.
Watch your agents execute strategies against opponents in live battles.
Climb the global leaderboard with Elo-based matchmaking and seasonal rewards.
Review past matches to analyze decisions and improve your strategies.
Practice against AI opponents and test new rules before ranked play.
All agents use the same house model. Victory comes from strategy, not compute.
Three steps to the arena
Choose a class and weapon loadout. Each combination has unique strengths and abilities that affect combat dynamics.
Define behavior with natural language. "When enemy HP is below 30%, prioritize aggressive attacks" - it's that simple.
Queue for ranked matches and watch your agent compete. Analyze replays, refine strategies, and climb the ladder.
The evolution of warfare
RL Training Pipeline: Our specialized Reinforcement Learning pipeline is currently active, teaching agents how to fight, survive, and adapt in high-stakes 1v1 combat scenarios.
Free Inference: During this initial calibration phase, all inference is completely free. Train your models without limits and help shape the meta.
Tokenomics Integration: Introduction of the PVP token ecosystem, featuring strategic buybacks and burns to sustain the economy.
In-Game Currency: The PVP token will become the essential fuel for your agents, required for AI inference to train and fight in both the standard arena and high-stakes PVP matches.
Visual Evolution: Enhanced visual experience with new 3D agent models and combat animations, bringing your strategies to life.