# vLLM serves DeepSeek-V4.1-Flash at 954K tokens/s per GPU on GB300 NVL72

> Original title: Serving DeepSeek-V4.1-Flash at Scale: vLLM Matches DeepInfer and Scales to GB300 NVL72

- Source: vLLM Blog
- Published: 2026-10-23T00:00:00.000Z
- Bad Influence: https://badinfluence.news/items/zyshay80si0tallpf7oeumvjj
- Original: https://vllm.ai/blog/2026-10-23-deepseek-v41-flash-pd-wide-ep

## Summary

vLLM reports 954K tokens/s per GPU at a P90 of 196 tokens/s per user on one GB300 NVL72 rack, using P/D disaggregation and Wide-EP, which it says is 7.9 times its earlier single-node result. The post says vLLM decode throughput is within 6% of DeepInfer, DeepSeek's own engine, across batch sizes 8 to 256, and attributes the gains to prefill, EPLB and CPU-overhead fixes.

## Why it matters

The post explains which measured bottlenecks were fixed and how each change moved throughput, useful for engineers tuning large MoE serving stacks.