Minimax M3 428B (23B active)
on 4× NVIDIA RTX Pro 6000 Blackwell · vLLM · 250,000 ctx
Oct 9, 2026
Use cases
long-contexttool-usevision
Summary
User reports MiniMax-M3 at ~113 tok/s single-stream decode on 4x RTX PRO 6000 Blackwell (96 GB each).
Setup is vLLM with the MXFP4 checkpoint (~256 GB, ~60 GB per card), BF16 KV cache, tensor parallelism 4, and 250K context; cold prefill runs ~2,800 tok/s flat from 8K to 131K.
Four concurrent requests reach ~184 tok/s aggregate. A second NVFP4/SGLang configuration on the same host decodes at ~87 tok/s single-stream with a ~643K-token KV pool and image/video support.