Skip to content
Live
Trending storyWatching

NVIDIA uses dual DGX Stations to serve GLM-5.3

1 reports1 outletsyesterdayUpdate

The story so far

AI overview

NVIDIA uses two DGX Stations with vLLM and tensor parallel processing to successfully run a 175B parameter GLM-5.3 NVFP4 model. The two DGX Stations are connected via ConnectX-8 RDMA.

AI-written from the reports · Updated yesterday

Latest14 Nov, 01:00
Running GLM-5.3 using two NVIDIA DGX Stations through vLLM

Coverage timeline

Follow the coverage to see every side of the story.

14 Nov
  1. NVIDIA Technical Blog
    Running GLM-5.3 using two NVIDIA DGX Stations through vLLM

    Two NVIDIA DGX Stations connected via ConnectX-8 RDMA, using vLLM and tensor parallel processing, successfully ran a 175B parameter GLM-5.3 NVFP4 model.

Heat over time

Now
5
Peak
1014 Nov, 02:00
24h change
–

Hover or tap the chart to see each hour; use the arrow keys on a keyboard.