The letter

One letter a week, starting soon.

The week's signal, what shipped, what is worth running tonight, and the editor's note. No tracking, no ads, nothing else.

We keep your address, your language and the date you joined, nothing else. Every letter has a one-click unsubscribe link that deletes the record.

← Accept All   Archive
vLLM BlogTools

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

11.12

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.

Models & releasesChips & computedeepseekperformancelarge-scale-servingdisaggregation
Read at vLLM Blog ↗

Related

More from vLLM Blog on Accept All.