OpenAI releases cybersecurity model GPT-6 Cyber: 51% relative gain on ExploitGym, uncovers 81 new Firefox vulnerabilities
OpenAI announced the launch of GPT-6 Cyber, a security-specific model, on November 5. It scored 48.5% on the ExploitGym benchmark, representing an approximate 51% relative improvement over GPT-5.6 Cyber's 32.1%.
Return to original text
A short introduction is retained here; the full text from the source website is not reposted. For specific wording and subsequent updates, please refer to the original article.
Open AIHOT original