Pinned
MiniMax M3: 428B params, 23B active per token, native video input, 1M context window. Open weights. We ran it and wrote down exactly what GPU config you need and what it costs. Save yourself the guesswork.
Article
How to run MiniMax M3 on Runpod
A 428B-parameter open-weights model just dropped with native video input, 1M token context, and frontier-level reasoning scores. That combination doesn't exist anywhere else in open weights right now....
