Microsoft Launches MAI-Code-2 and Code-2-Flash, Plus Cyber and Image Models

At Ignite on 17 November 2026, Microsoft announced MAI-Code-2 and MAI-Code-2-Flash, generally available in GitHub Copilot, plus MAI-Cyber-2-Flash and MAI-Image-2.7 in preview. Satya Nadella also announced MAI Licensed Weights, and GPT 6.2 and Meta models for Foundry.

Christopher Ort
By Christopher Ort · 2 min read
Share
A developer workstation with two monitors showing blurred code.
Image: AI-generated, i10X
On this page
  1. Quick Take
  2. In plain English
  3. What happened
  4. Background
  5. How it works
  6. Why it matters
  7. What's still open
  8. Sources
  9. i10x Perspective

Microsoft company page has launched a new family of its own coding models, MAI-Code-2 and MAI-Code-2-Flash, together with a security model and two image models. Satya Nadella announced the models in his Ignite keynote on 17 November 2026.

i10X Quick Take

Quick Take

  1. MAI-Code-2: flagship agentic coding model, generally available on GitHub Copilot in the model picker and in Copilot Fusion.
  2. MAI-Code-2-Flash: built for fast, high-volume work; generally available on GitHub, private preview in Foundry.
  3. MAI-Cyber-2-Flash: security model in preview, available only in Codename MDASH.
  4. MAI-Image-2.7 and 2.7-Flash: image generation and editing models in preview.
  5. Also announced: GPT 6.2 and Meta’s latest models coming to Foundry, plus MAI Licensed Weights for trusted customers.

In plain English

Microsoft is building its own AI models instead of only reselling others. The new coding models aim to do everyday company programming work at a lower price than frontier models. The Flash versions are smaller and quicker, for simple edits and explanations. Microsoft says the security model beats larger models at a fraction of the cost per task, which is its own claim.

What happened

Nadella said Code 2 “is at the frontier of cost-performance for everyday enterprise coding tasks” and “excels in taking code to production, deep debugging, multifile implementation, repo-wide reasoning and UI at a fraction of frontier cost.” Of the Flash model he said it “brings near-flagship quality for a much lower cost.” On security he said Cyber 2 Flash “is a flash-sized model that delivers performance ahead of frontier models at a fraction of the cost per task.”

Background

Microsoft’s MAI family is its in-house line, built alongside its partnerships with outside labs. Foundry now offers more than 11,000 models from OpenAI, Claude, Mistral and the open-source community, according to the Book of News. For a related development on the developer side, see our report on GitHub Copilot model deprecations.

How it works

Microsoft says it co-designs each model with the tools where it will run, for example Cyber 2 with its Perception security system. A model router in Foundry now draws on the top 50 models and supports AI Gateway. Nadella said MAI Licensed Weights means trusted customers will be able to get access to Microsoft AI model weights.

Why it matters

For developers, a cheaper Microsoft-built option appears directly in GitHub Copilot.

For enterprises, weights access could allow tuning inside their own boundaries.

For model vendors, Microsoft is building a competing first-party line while still hosting rivals.

What’s still open

  • Independent benchmark results; the keynote gave no numbers.
  • Pricing per token for the Code 2 models.
  • Who qualifies for MAI Licensed Weights and on what terms.

Sources

  • Microsoft, “Ignite 2026: Building the frontier ecosystem,” Satya Nadella, The Official Microsoft Blog, 17 November 2026
  • Microsoft, “Microsoft Ignite 2026 Book of News,” 17 November 2026

i10x Perspective

The performance claims come from Microsoft and are not yet independently tested. The strategic signal is stronger than the specs: Microsoft wants a first-party model for each job, sold at lower cost than the frontier. More in our Models section and Coding section.

Christopher Ort

Christopher Ort

Christopher Ort handles SEO, AI agent workflows and PR & partnerships at i10x.ai, the all-in-one AI workspace built by i10X Global Pte. Ltd. in Singapore. He is a graduate of Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU). He is the author of “The Style Penalty”, a study published on SSRN that analyses 1,576 evaluation decisions by four frontier AI models to show how AI-written CVs fare in automated hiring screens; the research has been covered by heise online and t3n.

Timeline

What happened before

  1. Release

    MAI-Code-1.1-Flash

    256K context

    microsoft.ai (source, opens in a new tab)
  2. Partnership

    Microsoft commits multiple billions to Mistral’s European GPU capacity

    With Microsoft · Compute

    reuters.com (source, opens in a new tab)
  3. Release

    MAI-Thinking-1 (7-model MAI family)

    Flagship reasoning model plus MAI-Code-1-Flash, MAI-Image-2.5, MAI-Transcribe-1.5 and MAI-Voice-2.

    microsoft.ai (source, opens in a new tab)
  4. Partnership

    Microsoft and OpenAI amend deal, ending exclusive licence

    With OpenAI · Cloud

    openai.com (source, opens in a new tab)
  5. Partnership

    Microsoft and NVIDIA to invest up to $15B in Anthropic

    With NVIDIA, Anthropic · Compute · $15B

    Microsoft (up to $5B) and NVIDIA (up to $10B) to invest in Anthropic; Anthropic commits to $30B of Azure compute.

    anthropic.com (source, opens in a new tab)
Full Microsoft timeline ›

Editor's picks

  1. Microsoft Teams Lets AI Agents Join as Digital Teammates With Their Own Identity
  2. Azure Cobalt 200 VMs Reach General Availability With Up to 50% More CPU Performance
  3. Microsoft Launches AI Transparency Initiative With NIST Standards Work and 40 Partners
  4. Agent 365 Gets Business Analytics and FinOps for AI as Nadella Cites 50 Million Agents
  5. Microsoft Perception Is Generally Available, With Red, Blue and Green Security Agents