CoreWeave, NVIDIA, and IBM Achieve Record MLPerf Results with GB200 Superchips

CoreWeave, in collaboration with NVIDIA and IBM, has submitted the largest-ever MLPerf Training v5.0 results using NVIDIA GB200 Grace Blackwell Superchips, achieving breakthrough performance on complex AI models.

CoreWeave, in collaboration with NVIDIA and IBM, has submitted the largest-ever MLPerf Training v5.0 results using NVIDIA GB200 Grace Blackwell Superchips, announced in a press release. The submission utilized 2,496 NVIDIA Blackwell GPUs on CoreWeave's AI-optimized cloud platform, marking the largest NVIDIA GB200 NVL72 cluster ever benchmarked under MLPerf.

The submission achieved a breakthrough result on the Llama 3.1 405B model, completing the run in just 27.3 minutes. This performance was more than twice as fast as other submissions with similar cluster sizes, highlighting the significant performance leap enabled by the GB200 NVL72 architecture.

CoreWeave's infrastructure demonstrated its capability to deliver consistent, high-performance AI workloads, reinforcing its leadership in supporting demanding AI tasks. These results translate to faster model development cycles and optimized costs for CoreWeave's customers, allowing them to efficiently scale and deploy AI models.

We hope you enjoyed this article.

Consider subscribing to one of several newsletters we publish like Silicon Brief.

Also, consider following us on social media:

AI Chips & Datacenters

AI Brief (X)

More from: Chips & Data Centers

Mindbeam AI Launches Litespark Framework on AWS

Jun 05

Vast.AI Opens New Office in San Francisco to Tap AI Talent

Jun 05

US Signal Invests $200 Million in Data Center and Fiber Expansion

Jun 05

CoreWeave Appoints Ernie Rogers as Chief Architect of Strategic Financing

Jun 05

CloudZero Partners with AWS for AI-Driven Cloud Efficiency

Jun 05

Subscribe to Silicon Brief

Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.

Market report

AI’s Time-to-Market Quagmire: Why Enterprises Struggle to Scale AI Innovation

ModelOp

The 2025 AI Governance Benchmark Report by ModelOp provides insights from 100 senior AI and data leaders across various industries, highlighting the challenges enterprises face in scaling AI initiatives. The report emphasizes the importance of AI governance and automation in overcoming fragmented systems and inconsistent practices, showcasing how early adoption correlates with faster deployment and stronger ROI.