Model Recipes
MiniMax

MiniMaxAI/MiniMax-M2

MiniMax M2 MoE language model (230B total / 10B active) for coding, agent toolchains, and long-context reasoning — native FP8 checkpoint, with an NVFP4 variant for Blackwell

Open-source MoE with strong SWE-Bench and Terminal-Bench performance, 196K context

moe230B / 10B196,608 ctxvLLM 0.11.0+text
Guide