CoreWeave, NVIDIA, And IBM Submit Record-Breaking MLPerf Results Using NVIDIA GB200 Grace Blackwell Superchips
In Brief
CoreWeave, in collaboration with NVIDIA and IBM, completed the largest-ever MLPerf Training v5.0 submission using 2,496 NVIDIA Blackwell GPUs on its AI-optimized cloud platform.
Software cloud platform CoreWeave announced that it has collaborated with NVIDIA and IBM to complete the largest MLPerf Training v5.0 submission to date using NVIDIA Blackwell technology. The effort utilized 2,496 NVIDIA Blackwell GPUs operating on CoreWeave’s cloud infrastructure, which is optimized for AI.
This benchmark represents the most extensive NVIDIA GB200 NVL72 cluster evaluated under MLPerf standards, measuring 34 times larger than the next largest submission from a cloud service provider. The result underscores the scalability and capability of the platform in supporting high-performance AI workloads.
The benchmark produced a notable outcome on the most advanced and resource-intensive model in the MLPerf suite—Llama 3.1 405B—completing the training task in 27.3 minutes. When evaluated against other submissions utilizing similarly sized clusters, the GB200 setup operated by CoreWeave demonstrated training speeds over twice as fast.
This performance reflects the efficiency of the GB200 NVL72 architecture and the ability of the underlying infrastructure to sustain high-level AI task execution. These results have practical implications, as they can lead to shorter development timelines and improved cost-efficiency in model training.
By enabling reduced training durations and effective workload scaling, the technology contributes to optimized resource usage. Additionally, with leading results in both MLPerf Inference v5.0 and Training v5.0, and as the only cloud platform in the Platinum tier of SemiAnalysis’ ClusterMAX, the platform demonstrates strong competitiveness and capability in AI infrastructure delivery.
CoreWeave Grows AI Cloud Capacity With New Data Center Deals And $375M Funding
CoreWeave operates as an AI-focused hyperscale cloud provider, offering advanced software infrastructure designed to support emerging developments in AI. Its platform enables accelerated computing capabilities for organizations and research institutions engaged in high-performance AI applications.
Recently, CoreWeave has entered into two long-term lease agreements with Applied Digital, securing 250 megawatts of IT load capacity at a data center campus located in Ellendale, North Dakota. The initial 100 megawatts are scheduled to become operational in the fourth quarter of 2025, with an additional 150-megawatt facility expected to launch by mid-2026. The company also retains the option to expand its capacity by up to 300 megawatts.
Following this development, its stock experienced a 25% increase, with an additional 5% rise in pre-market trading, resulting in a total gain of 276% so far this year.
In a separate move to support its growth, CoreWeave has secured $375 million in financing from SMBC to help advance its infrastructure expansion.
Disclaimer
In line with the Trust Project guidelines, please note that the information provided on this page is not intended to be and should not be interpreted as legal, tax, investment, financial, or any other form of advice. It is important to only invest what you can afford to lose and to seek independent financial advice if you have any doubts. For further information, we suggest referring to the terms and conditions as well as the help and support pages provided by the issuer or advertiser. MetaversePost is committed to accurate, unbiased reporting, but market conditions are subject to change without notice.
About The Author
Alisa, a dedicated journalist at the MPost, specializes in cryptocurrency, zero-knowledge proofs, investments, and the expansive realm of Web3. With a keen eye for emerging trends and technologies, she delivers comprehensive coverage to inform and engage readers in the ever-evolving landscape of digital finance.
More articles
Alisa, a dedicated journalist at the MPost, specializes in cryptocurrency, zero-knowledge proofs, investments, and the expansive realm of Web3. With a keen eye for emerging trends and technologies, she delivers comprehensive coverage to inform and engage readers in the ever-evolving landscape of digital finance.
