GPT-5.5-Cyber beats Anthropic's banned Mythos on CyberGym

Decrypt

OpenAI's GPT-5.5-Cyber scored 85.6% on the CyberGym benchmark, edging out Anthropic's Mythos 5 at 83.8%. Mythos was pulled offline after a U.S. export control directive, while GPT-5.5-Cyber remains available to trusted defenders.

OpenAIAnthropic

See 3 more sources

OpenAI is practically calling for regulation by the US government. Jokes aside: GPT Cyber ​​New...8 days agoKimmonismus

an updated GPT-5.5 Cyber outperforms Mythos 5 in CyberGym7 days agoOutside-Iron-8242Discuss

See, Dario? my GPT-5.5-Cyber beats your Mythos but I didn't go on an "existential-dread" press tour7 days agoItchy_Champion_86Discuss

Decrypt·Jose Antonio Lanz

Jun 23, 6:59 PM