AIToday
Large Language ModelsAI Coding AssistantsAI Business & IndustryPublickeyPublished: Nov 18, 2026, 10:01 JST

Microsoft unveils MAI Code 2, coding AI at 5% of Claude Opus 5.5 cost

Send on LINE
Microsoft unveils MAI Code 2, coding AI at 5% of Claude Opus 5.5 cost

3 Key Points

  1. What happened

    At Microsoft Ignite 2026, Microsoft announced MAI Code 2, which reaches 98% of GPT 6.1 Sol's and 96% of Claude Opus 5.5's benchmark performance at 40% and 5% of their cost.

  2. Why it matters

    Microsoft is pitching the models as delivering rival-level performance more cheaply, positioning them for everyday enterprise coding work at a fraction of competitors' prices.

WHO IT HITSEnterprise IT teams and developers choosing coding assistants can potentially cut AI tooling costs, as Microsoft claims comparable performance to leading models at 5–65% of competitor pricing.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Context & Analysis

Microsoft used its Ignite 2026 stage to frame the coding AI race around cost rather than raw capability. The company's own figures put MAI Code 2 within a few percentage points of GPT 6.1 Sol and Claude Opus 5.5 on a four-benchmark average of everyday enterprise coding tasks, but at 40% and 5% of those models' cost respectively.

The cheaper MAI Code 2 Flash follows the same logic for routine work. Microsoft says it matches GPT-6 Luna and Claude Haiku 5.5 on three coding benchmarks while costing 65% and 20% as much. That split — a flagship for production code and debugging, a low-cost model for explanation and small edits — suggests Microsoft is targeting the bulk of daily coding usage, where price per task may matter more than top-end performance.

Both models are aimed at enterprise coding work, a segment where Microsoft already has a large footprint through GitHub Copilot. The announcement did not include pricing or availability details beyond the performance and cost comparisons.

FAQ
What benchmarks were used to compare MAI Code 2?
Microsoft used four benchmarks covering everyday enterprise coding tasks, as seen in GitHub Copilot. The average showed MAI Code 2 at 98% of GPT 6.1 Sol and 96% of Claude Opus 5.5.
What is MAI Code 2 Flash designed for?
It is built for high-volume routine work such as code explanation, local edits, and routine fixes. Microsoft says it matches GPT-6 Luna and Claude Haiku 5.5 on three coding benchmarks at 65% and 20% of their cost.
Send on LINE

Also reported by SiliconANGLE AI

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleJordan Belfort's Tone AI sales startup hits $1 million revenue