François Chollet on GPT-6 Astra's ARC performance: 'GPT-6 Astra represents a step-function change in model capability for interactive reasoning problems. ... In fact, the continuous harness version significantly outperforms our human baseline in action efficiency across almost all levels. When we examined the reasoning chains to understand how the model operates, we found it performing highly efficient, on-the-fly symbolic world modeling for each game and level. It goes as far as developing…