GPT-6 Astra solves 49 of 50 physics research problems at $2.40 apiece
On the TheoPhysEval benchmark, GPT-6 Astra solved 49 of 50 unpublished theoretical physics calculations as accurately as leading specialists. In some cases where the agent appeared to fail, the experts' own reference solutions turned out to contain errors. Researchers say physics hasn't seen a breakthrough like those in mathematics because agents lack 'taste.'
