How AI is applied across API Evangelist and APIs.io. Read my AI disclosure →
API Evangelist API Evangelist
Discovery
Learnings
Guidance
Toolbox
Alignment
API Evangelist LLC

DeepSeek-V4.1-Flash on Vera Rubin NVL72: 58x Throughput vs. GB300 NVL72 at 387 tokens/s

calendar_today November 12, 2026 person Inferact and the vLLM Team domain vllm

On one NVIDIA Vera Rubin NVL72 rack, vLLM serves DeepSeek-V4.1-Flash on SemiAnalysis AgentX at 1.7M tokens/s per GPU at a P90 of 387 tokens/s per user, 58x GB300 NVL72 at the same interactivity.

open_in_new Read original post