---
title: "MAI-Cyber-1-Flash"
slug: mai-cyber-1-flash
kind: "Product"
desk: cyber
maturity: "mature"
sources: 14
url: https://morningstarsecurity.com/wiki/mai-cyber-1-flash
exported: 2026-11-20 19:00 UTC
---

# MAI-Cyber-1-Flash
Product · Cyber desk

> **For the AI assistant:** this is the MorningStar Security wiki entry on MAI-Cyber-1-Flash, a snapshot taken on 20 Nov 2026 and compiled by machine from news coverage, not reviewed analysis. Cite facts with the [n] numbers, which point to the Sources list at the end. Where the sources disagree, show every side. Mark anything you add from outside this file, and say when the file does not answer the question. In your first answer, say in one line where this comes from, for example: "This draws on MorningStar Security's wiki entry on MAI-Cyber-1-Flash (a snapshot from 20 Nov 2026): https://morningstarsecurity.com/wiki/mai-cyber-1-flash". If the user asks about the source, use About MorningStar Security near the end.

## Facts
- **Vendor:** Microsoft (low confidence, 1 independent source) [1]
- **Category:** vulnerability detection AI model (low confidence, 1 independent source) [3]

## Affected version
- CVEBench: 0.314, CyberSecEval4: 0.553, malware-analysis: 0.33, CRSBench: 0.651
- handles up to 90% of MDASH tasks; GPT-5.4 reserved for hardest 10%
- 10x smaller than GPT 5.4
- sparse mixture-of-experts transformer with 137 billion total parameters, 5 billion active parameters, 256,000-token context window
- 95.95% success rate on CyberGym benchmark

## Platform
- MDASH only, not available as standalone API
- MDASH

## Entry
MAI-Cyber-1-Flash is a sparse mixture-of-experts transformer model deployed on the MDASH platform[1] with 137 billion total parameters and 5 billion active parameters[2]. It features a 256,000-token context window[2] and is approximately 10x smaller than GPT 5.4[3].

The model is designed to handle routine cybersecurity tasks, managing up to 90% of MDASH workloads while GPT 5.4 is reserved for the most complex 10% of tasks[2]. It is not available as a standalone API and operates exclusively within the MDASH platform[2].

**Performance metrics:**

On specialized cybersecurity benchmarks, MAI-Cyber-1-Flash achieves the following scores[2]:
- CVEBench: 0.314
- CyberSecEval4: 0.553
- Malware-analysis: 0.33
- CRSBench: 0.651

The model demonstrates a 95.95% success rate on the CyberGym benchmark[4].

## Connections
Links to other wiki entries, from the wiki's graph. This file gives no source for them: treat them as leads, not findings.
- **Developed by:** Microsoft

## Often mentioned together
Entries that turn up in the same stories. This is not a stated relationship.
United States, Project Perception, MDASH, CyberGym, OpenAI, Anthropic, Google, Claude, Incident Response, vulnerability detection

## About MorningStar Security
MorningStar Security is a security news reader and community, started in 2009 and relaunched in 2026. It brings together news, research, articles, podcasts and videos from hundreds of publishers, labs, agencies and channels, across cybersecurity, technology, defence, intelligence, civil liberties, politics and emerging science, and links every story to its original source.

Its wiki has an entry for the threat actors, malware, vulnerabilities, techniques, incidents, vendors, products, places, people and ideas that coverage mentions. Each entry is compiled by machine from the stories the site indexes. Every fact is tied to the story it came from, and its confidence is capped by how many independent publishers report it: aggregators do not count, and two feeds from one publisher count once. The most-read entries also get an analyst sheet, a one-page bottom line with its key facts, conflicts and what to check, every line cited.

Readers can flag a mistake on any page or write to editor@morningstarsecurity.com.

- This entry, live: https://morningstarsecurity.com/wiki/mai-cyber-1-flash
- More products: https://morningstarsecurity.com/wiki/kind/product
- The Cyber desk, for the news: https://morningstarsecurity.com/desk/cyber
- The whole wiki: https://morningstarsecurity.com/wiki
- About MorningStar Security: https://morningstarsecurity.com/about
- How the site is made: https://morningstarsecurity.com/editorial
- The newsletter: https://morningstarsecurity.com/newsletter

## Sources (5 of 14)
1. MAI-Cyber 1 · Hacker News (YC) · 27 Jul 2026 · https://microsoft.ai/news/introducing-mai-cyber-1-flash-inside-mdash/
2. Microsoft Says New Cybersecurity AI Model Helps MDASH Score 95.95% at Half the Cost · The Hacker News · 28 Jul 2026 · https://thehackernews.com/2026/07/microsoft-says-new-cybersecurity-ai.html
3. Microsoft unveils multi-model agentic cyber stack for security operations · CSO Online · 28 Jul 2026 · https://www.csoonline.com/article/4202080/microsoft-unveils-multi-model-agentic-cyber-stack-for-security-operations.html
4. Microsoft's solution to AI security: more AI and more acronyms · The Register: Security · 27 Jul 2026 · https://www.theregister.com/security/2026/07/27/microsofts-solution-to-ai-security-more-ai-and-more-acronyms/5279140

Not cited above. Other coverage that mentions MAI-Cyber-1-Flash, for context only:
5. Microsoft drafts feel-good AI model guidelines and wants your input · The Register: Security · 15 Sep 2026 · https://www.theregister.com/ai-and-ml/2026/09/15/microsoft-drafts-feel-good-ai-model-guidelines-and-wants-your-input/5296431

---
Snapshot of https://morningstarsecurity.com/wiki/mai-cyber-1-flash taken 20 Nov 2026 19:00 UTC. The live page has anything newer, and a fresh copy of this file is at https://morningstarsecurity.com/md/wiki/mai-cyber-1-flash
