Show HN: Jev vs. GPT-5.6 and Claude Haiku at Pong

Jev, a typed-decision AI model, returns decisions in 227ms compared to 2.5-3.5 seconds for chat models like GPT-5.6 and Claude Haiku, demonstrating dramatically faster response times in a real-time Pong game benchmark. The performance difference is visualized through a simultaneous four-player Pong game where each model controls a paddle at its own speed, with Jev making significantly more decisions per game due to its lower latency. This comparison highlights how architectural differences between decision-focused and chat-based models impact performance in real-time applications.