Microsoft completes pre-training of MAI-Base-2 2.8T model

Microsoft AI announced completion of pre-training for MAI-Base-2, a mixture-of-experts foundation model with 2.8 trillion total parameters and 80 billion active parameters. Training used 16,000 NVIDIA GB200 GPUs over 38 days. Microsoft stated the model outperforms every open-weight model tested on held-out evaluations and is 4-12x more pre-training FLOPs-efficient than DeepSeek-V4-Pro-Base. The model is now in post-training with plans to license weights to trusted partners.