Broadcom Validates Top AI Models for On-Premises Deployment via VMware Cloud Foundation — Broadcom Inc.

Event summary

Broadcom announced validation of leading AI models from Google, NVIDIA, NEC, Alibaba Cloud, and Z.ai for deployment on VMware Cloud Foundation (VCF).
56% of enterprises plan to run production AI inferencing on private cloud, per Broadcom's Private Cloud Outlook 2026.
VCF supports mixed compute across AMD, Intel, and NVIDIA hardware, with vLLM as the default model runtime.
Models validated include NVIDIA's Nemotron 3, Google's Gemma 4, NEC's cotomi, Alibaba's Qwen 3.7-Max, and Z.ai's GLM 5.2.

The big picture

Broadcom's move underscores the growing enterprise demand for private AI clouds, driven by data sovereignty, security, and regulatory compliance. The validation of top AI models for VCF positions the platform as a key enabler for on-premises AI deployment at scale, particularly in industries with stringent governance requirements. This aligns with broader industry trends toward decentralized AI infrastructure and the rise of sovereign AI frameworks.

What we're watching

Adoption Pace: How quickly enterprises will migrate AI inference workloads from public to private clouds, driven by privacy and governance concerns.
Hardware Flexibility: Whether VCF's support for mixed compute across AMD, Intel, and NVIDIA will influence hardware purchasing decisions in enterprise AI deployments.
Model Ecosystem: The pace at which additional AI models will be validated for VCF, expanding the platform's appeal to enterprises with diverse AI needs.