Automatic failure detection and recovery on GB300
Company: Nebius
The claim, verbatim
Nebius infrastructure automatically detects node failures on NVIDIA GB300 NVL72 racks and returns them to service without manual intervention
Source (primary)
Nebius Rated Platinum in SemiAnalysis ClusterMAX™ 3.0 - Nebius (-, news_article)
View cached copy (2026-09-25)Live source ↗
Quote: “After a synthetic error was injected into an NVIDIA GB300 NVL72 rack, Nebius detected the failure immediately and automatically returned the node to service. No manual intervention was required.”
How we checked this
Checked on September 28, 2026. The cited source supports every part of this claim.
The post describes a SemiAnalysis test in which a synthetic error was injected into a GB300 NVL72 rack. Nebius detected the failure and automatically returned the node to service without manual intervention.
Confirmed in the source:
- Nebius detected a node failure on an NVIDIA GB300 NVL72 rack immediately
- Nebius automatically returned the failed node to service
- No manual intervention was required
What we did: Read our cached copy of the publisher (https://nebius.com/blog/posts/nebius-platinum-clustermax-3-0) in full (7,135 characters, retrieved September 25, 2026) and checked each assertion in the claim against it.
Additional evidence
confirms Nebius Rated Platinum in SemiAnalysis ClusterMAX™ 3.0 - Nebius
Quote: “After a synthetic error was injected into an NVIDIA GB300 NVL72 rack, Nebius detected the failure immediately and automatically returned the node to service. No manual intervention was required.”