From the source
Microsoft AI introduced MAI Cyber 2 Flash, a specialist security model built on
Microsoft's proprietary security data, which the company says matches GPT 6 Astra on real-world vulnerability discovery at less than 2% of the cost per task.
- The company says security teams need AI that can find vulnerabilities, recommend fixes and investigate threats around the clock, and that doing so at scale takes computing power whose cost adds up quickly.
- On CyberGym-E2E, MAI Cyber 2 Flash scored 92.3% task success against 91.5% for GPT 6 Astra using half the reasoning budget, and on WebExploitBench it scored 40.4% against 38.4% for GPT 5.6 Cyber at the same reasoning budget.
- The model has a one-million-token context window, four times larger than MAI Cyber 1.1 Flash, which Microsoft says lets it analyze more of a codebase during complex, multi-step investigations.
- Microsoft says early evaluations through Microsoft Perception show performance comparable to leading models on investigation agent and security detection authoring.
- Insight Enterprises reported a 20% token-efficiency advantage for MAI Cyber models versus GPT 5.4, GPT 5.3 Codex and GPT 5.4 Mini in its evaluations.
- MAI Cyber 2 Flash is coming exclusively to Microsoft security products, including Codename MDASH, where it will automatically replace the existing model in the MAI profile for eligible customers starting mid-December.






