← Monday, September 28, 2026InferenceBenchmark

DSX MaxLPS raises GPU throughput 49.2% within the same power budget

developer.nvidia.com by Sarah McKenney 9 min read Editor score 8/10

In an evaluation, DSX MaxLPS ran 192 GPUs instead of 140 within a 264.4 kW budget, raising aggregate throughput 49.2%; P99 time to first token rose 17%. The results show why capacity gains need to be weighed against tail latency.

What

DSX MaxLPS dynamically allocates power across participating resources under policy controls.

How

On GB300 NVL72 systems running Kimi K2.5, 192 GPUs raised normalized throughput 49.2% over a 140-GPU baseline at the same 264.4 kW.

Why it matters

P99 time to first token rose 17%, so operators must include tail latency when validating capacity gains.

Read on developer.nvidia.com →

Headline, summary, brief and diagram are written by a model from the source's own text; the link, byline and numbers come from the source. Read the original before relying on a claim.