AMD’s Helios Rack Reportedly Costs 40% More Than NVIDIA’s Vera Rubin, Yet Microsoft Hedges Its Bets By Committing To Deploy Both

wccftech | July 29, 2026, 6:35 p.m. | Original

Microsoft is moving aggressively to expand its compute footprint, and hedging its bets on the way by tapping rack-scale systems from both NVIDIA and AMD in what is an echo of its AI model-agnostic approach within Azure and Copilot.

Microsoft's CEO, Satya Nadella, has revealed a nugget during the just-concluded earnings call, declaring that Azure will be among the first cloud platforms to "deploy next generation rack scale AI infrastructure based on AMD Helios and NVIDIA Vera Rubin."

For the benefit of those who might not be aware, NVIDIA's Vera Rubin system consists of:

  1. NVIDIA Vera Rubin NVL72 Compute Tray (Rubin + Vera + Grace Bluefield + ConnectX-9)
  2. NVIDIA Vera Rubin NVL72 NVLink Switch Tray (NVLink6 Switch)
  3. NVIDIA Vera CPU Tray (Vera)
  4. NVIDIA Spectrum-6 SPX Switch Tray (Spectrum-6 Switch)
  5. NVIDIA Groq 3 LPX Tray (Groq 3)
  6. NVIDIA Vera Bluefield-4 STX Storage Tray (Vera Bluefield)

Moreover, as we explained in a dedicated post recently, Helios is AMD's first full-stack rack-level solution for AI workloads, consisting of:

  1. AMD Instinct MI455X GPUs
  2. 6th Gen AMD EPYC CPUs
  3. AMD Pensando AI Network Interface Cards (NICs)
  4. AMD Pensando DPU
  5. AMD Infinity Fabric
  6. AMD ROCm Software Stack

Do note that, according to Futurum, AMD will price the Helios at $5 million to $5.5 million per rack , which would constitute a premium of around 40 percent relative to the expected price of $3.5 million to $4 million for the second-gen NVIDIA Rubin rack.

Finally, by leveraging rack-scale systems from both NVIDIA and AMD, Microsoft appears to be replicating its AI model-agnostic strategy for Azure and Copilot.

According to The Information, Microsoft is already working to offer access to Moonshot's Kimi K3 model via its Azure cloud services. What's more, the Redmond giant is also evaluating the model's efficacy for deployment within Copilot.

This follows an Axios report back in June that suggested Microsoft might opt for DeeSeek's V4 model , or another similar one, albeit hosted on its own infrastructure, to power Copliot Cowork , which combines Copilot's enterprise-related functions with advanced AI models from OpenAI and Anthropic to make agentic AI-powered workflows a relative breeze.

Follow Wccftech on Google to get more of our news coverage in your feeds.

Intel Cracks The Encapsulation Wall Blocking Hyper-Large Chips, Pushing Advanced Packaging To 24x Reticle Size and Beyond Stupid Never Dies Takes On Square Enix On October 21, As It Lands Close To Final Fantasy Resonance