Our initial evaluation suggested that GPT-6-Astra's score was in the late eighties followed closely by Opus 5.5

But, as we started to investigate the mistakes that the models might have made, the models, especially Astra, insisted that they were right and that it was the expert-supplied answers that were erroneous. When we went back to the experts it turned out, in multiple instances, that that the experts found some subtle errors buried deep inside their calculations. As they revised their answers, Astra's score started to inch up and it finally ended up with 98% (49/50).

Also:

I don't know of any individual who can solve any significant fraction of our final collection of problems without AI assistance. For instance, I can solve the problems I contributed myself and understand a few others. Most problems don't even make sense to me.

Paper: https://arxiv.org/abs/2610.16733
Quote from an author summary: https://www.facebook.com/suvrat.raju/posts/pfbid0v6GWqNBXbdiQz1HEuTdYtfeio85nww5D1kU2unqkgUa8WtASeYyHePm91bZZt6YJl