Dell and NVIDIA have released a beast of an AI solution, combining PowerEdge servers with Blackwell Ultra GPUs that’ll make your current setup look like a calculator. This powerhouse delivers up to 4x faster language model training, supports a whopping 192 GPUs, and comes in both air and liquid-cooled flavors. Tech giants like Foxconn are already jumping on board, planning to deploy 10,000 NVIDIA GPUs. The future of AI infrastructure just got a serious upgrade.
Their expanded collaboration delivers a thorough AI solution that tackles the thorniest deployment challenges, from GPU management to resource utilization.
Think of it as the Swiss Army knife of AI infrastructure, but without the tiny scissors nobody knows how to use.
The star players in this AI dream team are Dell’s next-gen PowerEdge servers featuring NVIDIA’s Blackwell Ultra GPUs. Available in both air-cooled and liquid-cooled versions, these beefy machines can support up to 192 GPUs with direct-to-chip liquid cooling.
PowerEdge servers with Blackwell Ultra GPUs: computing beasts that handle up to 192 GPUs with cooling that would make a polar bear shiver.
For context, that’s enough computing power to simulate alternate universes where your fantasy football team actually wins.
Performance numbers don’t lie—these systems deliver up to 4x faster large language model training with 8-way NVIDIA HGX B300. The PowerEdge XE9680 has already become Dell’s fastest-ramping solution ever, and these new models are its beefier successors.
What truly elevates this partnership is the integration of NVIDIA Run:ai platform, providing orchestration capabilities that guarantee GPU resources aren’t sitting idle like that fancy kitchen gadget you bought and never use.
Even Foxconn has jumped on board, planning to deploy 10,000 NVIDIA GPUs through its Big Innovation Company as an NVIDIA Cloud Partner.
That’s the kind of scale that makes regular data centers look like calculator watches from the ’80s.
For enterprises stuck in AI experimentation purgatory, Dell and NVIDIA have created a streamlined path to full implementation. These innovations align with the global AI market projections showing growth to $1.85 trillion by 2030.
Their joint solution spans the entire AI lifecycle—from model development to inference—aligning AI operations with business objectives without adding unnecessary complexity. The new NVLink Fusion technology further enhances connectivity, allowing cloud providers to scale their AI infrastructure to millions of GPUs.
The AI arms race just got more interesting, and Dell and NVIDIA are clearly not playing for second place.
