Big news！ Our new MAI-Cyber-1-Flash model combined with MDASH， our multi agent security harness， delivers 96% on the CyberGym benchmark， 12pts above Mythos， at HALF the cost.
Proud of the team. More details in THREAD：
Mustafa Suleyman@mustafasuleyman · XAI 摘要
重大消息！我们的新 MAI-Cyber-1-Flash 模型结合 MDASH（多智能体安全框架），在 CyberGym 基准测试中达到 96%，比 Mythos 高出 12 个百分点，成本仅为其一半。为团队感到骄傲。更多详情见 THREAD：
Big news！ Our new MAI-Cyber-1-Flash model combined with MDASH， our multi agent security harness， delivers 96% on the CyberGym benchmark， 12pts above Mythos， at HALF the cost.
Proud of the team. More details in THREAD：