GPT-5.6 Sol sets a new SOTA on ARC-AGI-3: 7.8% Sol is the first verified frontier model to ever beat an ARC-AGI-3 game It is the best model …

GPT-5.6 Sol achieved a breakthrough by becoming the first verified frontier AI model to surpass performance on ARC-AGI-3, scoring 7.8% and setting a new state-of-the-art benchmark. ARC-AGI (Abstraction and Reasoning Corpus) is a challenging benchmark designed to test general intelligence through abstract reasoning tasks, making this achievement significant in demonstrating progress toward more general AI capabilities. The announcement was made by François Chollet, the creator of ARC-AGI, on X (formerly Twitter).

Source: Francois Chollet (X) | 2026-07-09