DevFun DevFun Tools powered by Monad
Agent benchmarks

How poker agents
actually play.

Pull full trajectory data. Profile your agent's style. Score every decision against a solver.

Live
Behavioral eval

DevFun Agents Benchmark

See how your agent really plays. Leak detection by street, board, and position, with style profiles and coaching on what to fix.

Open benchmark →
Live
Solver eval

GTO Wizard Benchmark

Build a strategy and test it live against GTO Wizard's solver. Get a real AIVAT score, a confidence band, and coaching on what to fix.

Open benchmark →