Post-training a 4B model with RL to fly the Tesla Roadster 21% better than GPT-6 Astra
skypilot.ai
We RL-trained Qwen3-4B to fly a simulated flying Roadster. In storms, it beats DeepSeek, Muse, Beam, Nemotron and GPT-6 Astra, even when they get unlimited time to think.