Microsoft MAI-Thinking-1: A Fully Homegrown Reasoning Model Scoring 97% on AIME 2025
At Build 2026 on June 2, 2026, Microsoft officially unveiled MAI-Thinking-1 — the company's first fully in-house reasoning model, built without relying on distillation from OpenAI or DeepSeek. Scoring 97.0% on AIME 2025, it marks Microsoft's formal entry as an independent player in foundation model development.
Model Specifications
| Spec | Details |
|---|---|
| Developer | Microsoft |
| Architecture | Sparse Mixture of Experts (MoE) |
| Total Parameters | ~1 trillion (1T) |
| Active Parameters | 35 billion (35B) |
| Context Window | 256K tokens |
| Training Data | 30T tokens (50%+ code) |
| Training Hardware | 8,000 GB200 GPUs |
| Distillation | ❌ Fully in-house, no third-party distillation |
| Release Date | June 2, 2026 (Build 2026) |
Benchmark Performance
| Benchmark | MAI-Thinking-1 | Claude Sonnet 4.6 | DeepSeek V3.2 | Claude Opus 4.6 |
|---|---|---|---|---|
| AIME 2025 | 97.0% | 95.6% | 93.1% | 99.8% |
| AIME 2026 | 94.5% | — | — | — |
| SWE-Bench Pro | 52.8% | — | — | — |
| LiveCodeBench v6 | 87.7% | — | — | — |
On AIME 2025, MAI-Thinking-1 scored 97.0%, surpassing Claude Sonnet 4.6 (95.6%) and DeepSeek V3.2 (93.1%) — second only to Claude Opus 4.6 (99.8%).
It's worth noting that these are Microsoft's self-reported numbers and have yet to be independently reproduced and verified.
The Full MAI Model Family
Microsoft released seven MAI models at Build 2026:
| Model | Type | Description |
|---|---|---|
| MAI-Thinking-1 | Reasoning | Flagship reasoning model (the focus of this article) |
| MAI-Code-1-Flash | Coding | Fast coding model powering Copilot |
| MAI-Image-2.5 | Image Generation | Image generation model |
| MAI-Image-2.5 Flash | Image Generation | Fast image generation |
| MAI-Voice-2 | Voice | Voice model |
| MAI-Voice-2 Flash | Voice | Fast voice model |
| MAI-Transcribe-1.5 | Transcription | Speech-to-text model |
Why This Matters
1. Microsoft Breaks Free from OpenAI Dependency
In April 2026, Microsoft and OpenAI revised their partnership agreement, allowing Microsoft to develop its own foundation models. The launch of MAI-Thinking-1 signals that Microsoft has officially transitioned from being an OpenAI technology distributor to an independent competitor.
2. Trained on a Massive GB200 GPU Cluster
MAI-Thinking-1 was trained on 8,000 GB200 GPUs — one of the largest known GB200 cluster training runs to date. GB200 is NVIDIA's latest flagship GPU, underscoring the enormous hardware investment Microsoft has made.
3. Over 50% Code in Training Data
With more than 50% of its training data consisting of code, it's no surprise that MAI-Thinking-1 delivers strong results on SWE-Bench Pro (52.8%) and LiveCodeBench (87.7%). For coding-intensive use cases, this model could be a highly competitive option.
Recommended Use Cases
| Use Case | Recommended Model | Rationale |
|---|---|---|
| Mathematical reasoning (AIME-style) | MAI-Thinking-1 | 97.0% on AIME 2025, second only to Opus 4.6 |
| Everyday coding assistance | MAI-Code-1-Flash | Copilot-integrated, optimized for speed |
| Complex reasoning and architecture design | Claude Opus 4.8 | Strongest all-around capability |
| Large-scale batch processing | MAI-Thinking-1 | MoE architecture enables efficient inference |
| Data-sensitive environments | MAI-Thinking-1 | Azure-hosted with enterprise-grade security |
The Big Picture
The launch of MAI-Thinking-1 represents a fundamental shift in Microsoft's AI strategy.
Key takeaways:
- First fully in-house reasoning model — no distillation from OpenAI or DeepSeek
- 97.0% on AIME 2025 — beating Claude Sonnet 4.6, trailing only Opus 4.6
- Trained on 8,000 GB200 GPUs — one of the largest GB200 clusters in existence
- Seven MAI models launched simultaneously — spanning reasoning, coding, image generation, voice, and transcription
- Microsoft is now an independent AI competitor — no longer just OpenAI's distribution partner
For enterprises, MAI-Thinking-1 introduces a new option: a Microsoft-backed, Azure-hosted reasoning model with enterprise-grade security. As Microsoft integrates MAI models into Copilot and Azure AI, their real-world impact will become increasingly clear.
Loading...