LaunchAI ModelsOctober 23, 2026

MiniMax open-sources 428B-parameter M3.1 Flash weights

Read original source →technode.com

The mixture-of-experts model activates 23B parameters per token and handles text, image and video with a 1M-token context window. Weights follow the October 19 service launch, with deployment guidance for SGLang and vLLM plus multi-chip support via FlagOS.

1 source

More stories today

Open the live feed