minimax-m3.1POST
MiniMax M3.1 Flash open weights: 428B total, 23B active per token
MiniMax released M3.1 Flash weights at 428B total parameters with only 23B active per token, downloadable on Hugging Face and ModelScope; the poster sees sparse activation keeping inference costs down.
矿工阿三
@tarsixseveneig1Archived 3 days agoOriginally posted 2026.10.25
矿工阿三 @tarsixseveneig1Originally posted 2026.10.25
MiniMax把M3.1 Flash的权重公开了，总参数428B、每token激活23B，Hugging Face和ModelScope都能下。 这个尺寸敢直接放权重，配上稀疏激活的推理成本，诚意摆在这了。 428B的盘子每token只动23B，下载链接已经挂出来了。
x.com/tarsixseveneig1/status/2114084806141399266Archived 2026.10.25 12:06