Moe

Microsoft unveils MAI-Base-2 MoE with 2.8T total, 80B active parameters: Microsoft unveiled MAI-Base-2, a mixture-of-experts model with 2.8T total and 80B active parameters. It claims 4x-12x more pretraining FLOPs-efficiency than DeepSeek-V4-Pro-Base. Microsoft also cited MAI-Thinking-1, a general-purpose foundation model post-trained on MAI-Base-1 announced in June.

AI Models • Techmeme

Nature Will unveils Natura-1, a 50B MoE pretrained on 1M hours of human video: Nature Will's Natura-1 is a 50-billion-parameter MoE pretrained on one million hours of first-person human video. The model was then adapted with limited robot data to control 20-DoF robot hands.

Robotics • Pandaily

ModelBest open-sources MiniCPM-V 4.7, saying its MoE model runs in about 3GB on an iPhone 17 Pro Max: ModelBest open-sourced MiniCPM-V 4.7, its first MoE multimodal model, in 35B-A3B and 1B sizes. The company says the MoE version runs in about 3GB of memory on an iPhone 17 Pro Max and decodes at over 13 tokens per second.

open source • Pandaily