vLLM adds day-0 support for MiniMax-M3.1-Flash on NVIDIA and AMD GPUs
AIvLLM says it supports MiniMax-M3.1-Flash on day 0 across NVIDIA Blackwell and Rubin GPUs and AMD MI350-series GPUs. The post reports up to 1.7× faster indexer QK decode on Blackwell and 1.3× faster sparse attention decode with an NVFP4 KV cache.
AIWhy it matters.
The post gives concrete kernel, memory, and throughput figures for running MiniMax-M3.1-Flash on several GPU platforms, useful for teams planning inference deployments.
Source details
vLLM Blog · Inferact and the vLLM Team · vllm.aiPublished · added here
Original title: vLLM Delivers Day-0 Support for MiniMax-M3.1-Flash