Report
Microsoft unveils MAI-Base-2, a 2.8T-parameter mixture-of-experts model
Techmeme says it has 80B active parameters and relays Microsoft's claim of 4x–12x more pretraining FLOPs-efficiency than DeepSeek-V4-Pro-Base.
TLDR
Techmeme, linking to Microsoft AI, reports that Microsoft unveiled MAI-Base-2, a mixture-of-experts model with 2.8 trillion total parameters and 80 billion active parameters. Microsoft claims 4x–12x more pretraining FLOPs-efficiency than DeepSeek-V4-Pro-Base.
Combined views
1.1K
1 Source, first seen ago