Today for AI
HOT RADAR
computeHEAT 10.8°

AI CURATED HOT STORY · 11/12/2026

DeepSeek-V4.1-Flash Hits 58x Throughput on Vera Rubin NVL72

1 reports archived1 independent sourcesupdated 11/12/2026, 08:00:00
Synthesis & Latest Updates
1 Sources Cross-Validated

The vLLM team deployed DeepSeek-V4.1-Flash on a single NVIDIA Vera Rubin NVL72 rack, reaching 1.7M tokens/s per GPU and 387 tokens/s per user, a 58x gain over GB300 NVL72, with peak throughput of 2.52M tokens/s. The speedup stems from kernels tuned for Rubin's split memory and FP4 tensor cores, specialized low-latency decode kernels, and rack-wide KV cache and Engram tables shared over NVLink.

LATEST/The vLLM team deployed DeepSeek-V4.1-Flash on a single NVIDIA Vera Rubin NVL72 rack, reaching 1.7M tokens/s per GPU and 387 tokens/s per user, a 58x gain over GB300 NVL72, with peak throughput of 2.52M tokens/s. The speedup stems from kernels tuned for Rubin's split memory and FP4 tensor cores, specialized low-latency decode kernels, and rack-wide KV cache and Engram tables shared over NVLink.

Hourly heat curve

4 fully observed hours
Now
10.8
Peak
11.807:00
24h change
–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

Coverage timeline

Total 1 reports · Latest first
  1. vLLM 官方博客T1·78 pts

    DeepSeek-V4.1-Flash Hits 58x Throughput on Vera Rubin NVL72

    • A single Vera Rubin NVL72 rack sustains 1.7M tokens/s per GPU for DeepSeek-V4.1-Flash at 387 tokens/s per user, 58x over GB300 NVL72
    • Peak throughput reaches 2.52M tokens/s, with up to 1,810 tokens/s for a single user
    • Gains come from kernels tuned for Rubin's split memory and FP4 tensor cores, low-latency decode kernels, and rack-wide KV cache and Engram tables over NVLink