Skip to content
Trending storyDeveloping

NVIDIA submits Rubin NVL72 Kimi K3 benchmark result above 850 tok/s per user

1 article1 sourcesince Oct 27Last article

Overview

AISummary of 1 article

NVIDIA has submitted its first Rubin NVL72 benchmark result for Kimi K3, according to SemiAnalysis, which reports the system reaches over 850 tokens per second per user at 2.8 trillion parameters.

SemiAnalysis also reports more than 3x higher throughput on vLLM. The source says Rubin Software 1.0 general availability is coming soon. This is a single report from SemiAnalysis's post on X, and the figures have not been independently verified.

Written by AI from the articles below · updated Oct 27, 1:04 PM ET

Check the sources:

Latest developmentOct 27, 1:01 PM ET
NVIDIA submits first Rubin NVL72 Kimi K3 result at 850+ tok/s/user

Article timeline

The articles in this story. Times are ET.

Oct 27
  1. SemiAnalysisX
    NVIDIA submits first Rubin NVL72 Kimi K3 result at 850+ tok/s/user

    AIBREAKING NEWS🚨: NVIDIA just submitted its first Rubin NVL72 Kimi K3 submission, and it can reach over 850 tok/s/user, even at 2.8 Trillion parameters, with over 3x higher throughput on @vllm_project! Great work by the NVIDIA team, and excited for Rubin Software 1.0 GA coming soon!

    Image from @SemiAnalysis_'s post

Heat Trend

Current comparable heat
12
Peak comparable heat
20Oct 27, 2:00 PM ET
Change in past 24 hours
–

The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.