Broadcom Inc. (NASDAQ: AVGO) announced that AI models from Google, Nvidia, NEC, Alibaba Cloud, and Z.ai have been validated to run on VMware Cloud Foundation (VCF), according to a company press release issued at VMware Explore 2026 in Las Vegas.

The validated models include Nvidia's Nemotron 3, Google DeepMind's Gemma 4, NEC's cotomi, Alibaba Cloud's Qwen 3.7-Max, and Z.ai's GLM 5.2. Broadcom said VCF allows enterprises to deploy these models on-premises and offer them as a managed service to internal users.

Broadcom cited its Private Cloud Outlook 2026 report, which found that 56% of enterprises are already running or planning to run AI inferencing workloads on private cloud infrastructure. The company said VCF supports inference workloads, agentic applications, containerized services, and traditional virtual machines on a single platform.

Independent benchmark testing under MLPerf Inference v5.1 standards found that VCF delivers performance comparable to bare metal deployments, Broadcom said. The platform supports mixed compute hardware from AMD, Intel, and Nvidia, and uses vLLM as the default model runtime, which Broadcom said enables support for more than 150 open-source models.

"Broadcom is committed to giving enterprises the broadest set of AI models for their on-premises infrastructure, all validated on VCF," said Chris Wolf, global head of AI and advanced services, VMware Cloud Foundation Division, Broadcom.

NEC's cotomi model is described as optimized for Japanese language processing, with a claimed 40% improvement in token efficiency. NEC Chief AI Officer Akio Yamada noted that NEC also offers a private cloud service hosted in NEC data centers using VMware Cloud Foundation.