Microsoft AI unveils MAI-Base-2, a 2.8-trillion-parameter base model

Microsoft AI detailed MAI-Base-2, a mixture-of-experts base model with 2.8 trillion total and 80 billion active parameters, trained on 16,000 GB200 GPUs over 38 days, which it says is 4 to 12 times more pre-training-efficient than DeepSeek-V4-Pro-Base. The model, with native vision, is still in post-training.