Seven system configurations for MLPerf
Company: Nebius
The claim, verbatim
Nebius submitted seven system configurations including rack-scale NVIDIA GB300 NVL72, single-node HGX B300 and B200, 8x RTX PRO 6000, and Nebius VR200 NVL72
Source (primary)
MLPerf® Inference v6.1 Results: NVIDIA Vera Rubin NVL72 - Nebius (-, news_article)
View cached copy (2026-09-21)Live source ↗
Quote: “Seven system configurations span a wide range of inference hardware: the rack-scale Nebius system with NVIDIA GB300 NVL72 at three cluster sizes, single-node Nebius system with NVIDIA HGX B300 and NVIDIA HGX B200, a cost-efficient node with 8 NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, and the Nebius VR200 NVL72 preview.”
How we checked this
Checked on September 25, 2026. The cited source supports every part of this claim.
The post lists seven configurations covering exactly these systems, with the GB300 NVL72 run at three cluster sizes.
Confirmed in the source:
- Nebius submitted seven system configurations
- The configurations included a rack-scale NVIDIA GB300 NVL72 system
- The configurations included single-node NVIDIA HGX B300 and HGX B200 systems
- The configurations included a node with 8 NVIDIA RTX PRO 6000 GPUs
- The configurations included the Nebius VR200 NVL72
What we did: Read our cached copy of the publisher (https://nebius.com/blog/posts/mlperf-inference-v6-1-results) in full (19,885 characters, retrieved September 21, 2026) and checked each assertion in the claim against it.
Additional evidence
confirms MLPerf® Inference v6.1 Results: NVIDIA Vera Rubin NVL72 - Nebius
Quote: “Seven system configurations span a wide range of inference hardware: the rack-scale Nebius system with NVIDIA GB300 NVL72 at three cluster sizes, single-node Nebius system with NVIDIA HGX B300 and NVIDIA HGX B200, a cost-efficient node with 8 NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, and the Nebius VR200 NVL72 preview.”