Taking a look at the benchmarks that head the article, one thing becomes clear: it's not a Mythos-level model. It simply follows the continuationist saga we've seen with GPT 5.1, GPT 5.2, GPT 5.4 and now GPT 5.5 So a priori, only the jump in FrontierMath Tier 4 gets me excited

→ View original post on X — @dotcsv,