# vLLM adds day-0 support for MiniMax-M3.1-Flash on NVIDIA and AMD GPUs

> Original title: vLLM Delivers Day-0 Support for MiniMax-M3.1-Flash

- Source: vLLM Blog
- Published: 2026-10-23T00:00:00.000Z
- Bad Influence: https://badinfluence.news/items/id7c7wq4drb0dklndhygtub0l
- Original: https://vllm.ai/blog/2026-10-23-minimax-m31-flash

## Summary

vLLM 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. Prebuilt images are available, and the model reaches the nightly PyPI wheel after upstream merge, usually about a week later.

## Why 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.