---
format: "aidr-story-markdown/v1"
id: "ea8a00442d5df36af3ff34d67b363d29108ec13364ec20eccc3cd2c6f60fe346"
canonical_url: "https://aidr.today/ea8a0044?lang=en"
title: "Microsoft unveils MAI-Base-2, a MoE model with 2.8T total and 80B active parameters, claiming 4x-12x more pretraining FLOPs-efficiency than DeepSeek-V4-Pro-Base (Microsoft AI)"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-16T02:20:01.000Z"
category: "Models"
topics: ["microsoft","llm"]
source_urls: ["https://www.techmeme.com/261015/p59"]
summary: "Microsoft AI unveiled MAI-Base-2, a mixture-of-experts model with 2.8 trillion total parameters and 80 billion active parameters, claiming 4x-12x greater pretraining FLOPs-efficiency than DeepSeek-V4-Pro-Base. The announcement follows the June release of MAI-Thinking-1, a general-purpose foundation model built on MAI-Base-1."
---

# Microsoft unveils MAI\-Base\-2, a MoE model with 2\.8T total and 80B active parameters, claiming 4x\-12x more pretraining FLOPs\-efficiency than DeepSeek\-V4\-Pro\-Base \(Microsoft AI\)

> [Open the canonical story](<https://aidr.today/ea8a0044?lang=en>)

**Published:** 2026-10-16T02:20:01.000Z
**Category:** Models
**Topics:** microsoft, llm

## Summary

Microsoft AI unveiled MAI\-Base\-2, a mixture\-of\-experts model with 2\.8 trillion total parameters and 80 billion active parameters, claiming 4x\-12x greater pretraining FLOPs\-efficiency than DeepSeek\-V4\-Pro\-Base\. The announcement follows the June release of MAI\-Thinking\-1, a general\-purpose foundation model built on MAI\-Base\-1\.

## Sources

- [Story source](<https://www.techmeme.com/261015/p59>)