{
  "id": 14846245,
  "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)",
  "url": "https://urgent.news/2026/10/16/microsoft-unveils-mai-base-2-a-moe-model-with-2-8t-total-and-80b",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-16T02:20:01.000Z",
  "source": {
    "name": "Techmeme",
    "slug": "techmeme",
    "url": "https://microsoft.ai/news/frontier-pre-training-at-microsoft-ai/"
  },
  "original_language": "en",
  "account": "On June 1st, Microsoft unveiled MAI-Base-2, a large language model boasting an astounding 2.8 trillion total parameters and 80 billion active parameters. This model claims to be 4-12 times more efficient in terms of pre-training FLOPs compared to DeepSeek-V4-Pro-Base.\n\nMAI-Base-2 is an MoE foundation model, having been trained on top of MAI-Base-1, Microsoft's previous state-of-the-art general-purpose foundation model. The new model outperforms every open-weight model in Microsoft's testing across various held-out evaluations, with a lower bits-per-byte (BPB) score for text and code. For STEM, math, and Q&A tasks, evaluations were commissioned from vendors to ensure data freshness.\n\nWhat sets MAI-Base-2 apart is Microsoft's full control over the model pipeline, from data provenance to training compute. This allows the company to offer a unique, trusted model to its customers via the Foundry platform, which provides enterprises with access to high-quality intelligence.\n\nThe model's impressive efficiency is due to a combination of factors, including the use of 16k GB200 GPUs for 38 days of training and a sparse, well-chosen dataset. Through a scaling ladder approach, Microsoft was able to accurately predict MAI-Base-2's performance and demonstrate a 4.2-12.3x improvement in FLOPs-efficiency over DeepSeek-V4-Pro-Base.",
  "summary": "In June, we announced MAI-Thinking-1, a general-purpose foundation model post-trained on top of MAI-Base-1 with strong reasoning performance.",
  "key_points": [
    "MAI-Base-2 model unveiled with 2.8T total and 80B active parameters",
    "4-12 times more efficient in pre-training FLOPs than DeepSeek-V4-Pro-Base",
    "Model trained on MAI-Base-1, outperforms open-weight models in tests"
  ],
  "editors_take": "By controlling the entire pipeline from data to compute, Microsoft is positioning its high-efficiency mixture-of-experts models as a more trusted and predictable intelligence source for enterprise customers.",
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}