Microsoft

MAI-Base-2

In post-training

Microsoft AI said on Oct. 15, 2026 that pre-training of MAI-Base-2 is complete. It is a mixture-of-experts model with 80 billion active parameters and 2.8 trillion total parameters, trained on 16,000 Nvidia GB200 GPUs for 38 days in August and September 2026. Microsoft says the model is in post-training and that, on held-out tests it ran, the model beat every open-weight model compared. The 4.2 to 12.3 times FLOPs-efficiency claim versus DeepSeek-V4-Pro-Base is Microsoft's.

Latest update Oct 16, 2026

01

Current status

Pre-training announced complete on Oct. 15, 2026. No public weights or release date.

02

In the news1

  1. Microsoft says MAI-Base-2 pre-training is complete

    On Oct. 15, 2026 Microsoft AI said it finished pre-training MAI-Base-2, a mixture-of-experts model with 80 billion active parameters and 2.8 trillion total parameters. Microsoft said the run used 16,000 GB200 GPUs for 38 days in August and September 2026, and that on its own held-out tests the model beat every open-weight model it compared, with a claimed 4.2 to 12.3 times FLOPs-efficiency advantage over DeepSeek-V4-Pro-Base. The model is in post-training, with no release date or public weights.

    microsoft.ai

03

More from Microsoft4

04

More in Big Tech