A new GitHub project pits reinforcement learning algorithms against each other in a simulated car navigation task. Using PyBullet and Stable-Baselines3, SAC emerges as the clear winner in training efficiency and navigation success while A2C trails significantly behind.