MiniMax M3.1 Flash open weights: 428B total/23B active, 1M context

Wes Roth @WesRothOriginally posted 2026.10.26

Open original on X

MiniMax has released the weights for M3.1 Flash on Hugging Face and ModelScope. The multimodal model has 428 billion total parameters, with 23 billion activated per token. It accepts text, images and video, with a context window of up to one million tokens. MiniMax reports 73.7% on DeepSWE v1.1 and 61.7 on OSWorld 2.0. The weights are distributed under the MiniMax Community License.

This post quotes another post: “MiniMax M3.1 Flash open weights: 428B total, 23B activated”