OpenAI's GPT-5.5-Cyber beats Anthropic's banned Mythos model

GPT-5.5-Cyber scored 85.6% on the CyberGym benchmark, outperforming Anthropic's Mythos 5 (83.8%) and Claude Opus 4.7 (73.1%). Anthropic's Mythos was pulled offline on June 12 due to a U.S. government export ban over national security concerns.