Key Highlights
- Advanced Micro Devices shares climbed approximately 5% following news of strengthened Microsoft Azure ties
- Azure will integrate AMD Helios rack-scale systems for powering AI inference applications
- Helios integrates MI455X GPUs, EPYC Venice processors, Pensando networking technology and ROCm software stack
- Two additional EPYC-based VM series coming to Azure, plus broader Pensando DPU integration
- Early Helios adopters include Meta, OpenAI, Oracle, and now Microsoft
Shares of Advanced Micro Devices climbed approximately 5% during Monday trading after the chipmaker unveiled a significant deepening of its collaboration with Microsoft Azure.
Advanced Micro Devices, Inc., AMD
The centerpiece of this expanded partnership is AMD Helios — the company’s inaugural rack-scale artificial intelligence system. Azure will leverage Helios to deliver frontier AI model inference capabilities throughout its cloud services and for enterprise customers.
According to AMD, Helios represents a fully integrated solution combining Instinct MI455X graphics processors, EPYC Venice central processors, Pensando networking infrastructure, and ROCm software — all engineered specifically for large-scale AI training and inference operations.
Shipments of Helios to customers, Microsoft included, are scheduled to commence during the latter half of 2026.
The collaboration extends beyond Helios. Azure plans to introduce two additional virtual machine families utilizing 6th Generation AMD EPYC Venice chips. Azure HDv2 will serve agentic AI applications and data processing pipelines, while Azure HXv2 targets semiconductor design computing requirements.
Network infrastructure also factors into the expanded agreement. Microsoft and AMD are combining Azure Boost technology with AMD’s Pensando data processing units to enhance networking throughput and efficiency throughout cloud operations.
Direct Challenge to Nvidia’s Data Center Dominance
With Helios, AMD launches its most comprehensive offensive against Nvidia’s stronghold in data center artificial intelligence markets. Though Nvidia maintains market leadership in AI accelerator deployments, AMD is countering with an end-to-end offering encompassing processors, connectivity, and software tools.
Microsoft CEO Satya Nadella emphasized customer flexibility. “We are expanding the Azure infrastructure portfolio with AMD Helios to give customers the performance, scale and choice they need to build and run the next generation of AI applications,” he said.
AMD CEO Lisa Su reinforced the strategic importance, describing Microsoft’s commitment to deploy AMD solutions as “an important milestone” for scaling next-generation artificial intelligence infrastructure.
Major Tech Players Already Committed
Microsoft joins an already impressive roster of Helios customers. CNBC reports that Meta, OpenAI, and Oracle have secured early access to the platform.
This customer lineup positions AMD’s rack-scale offering with some of the most demanding AI computation consumers globally.
The broadened partnership additionally provides Azure customers with AMD-powered computing resources via Azure Foundry Managed Compute, enabling enterprise organizations to deploy and scale production artificial intelligence applications.
Microsoft shares declined roughly 1% following the announcement.
Both companies confirmed that AMD will initiate Helios deliveries to customers beginning in the second half of 2026.





