Introduction
NVIDIA AI Virtual Workstation (vWS) utilizes NVIDIA RTX Virtual Workstation, part of the NVIDIA vGPU software suite, along with data center GPUs to facilitate remote access to GPU acceleration for AI development. Thus, it serves as an excellent starting point for organizations embarking on their AI journey. AI vWS Toolkits are designed explicitly for diverse use cases and provide essential reference documentation and instructions to assist organizations in jumpstarting AI development. This toolkit includes the following guide on how to deploy NVIDIA’s Video Search and Summarization blueprint in a vWS environment. This blueprint utilizes a visual learning model to scan a video for a prompted event and alert when the conditions are met. Utilizing a Large Language Model (LLM), it also allows for inference for further details about each event. A concise walkthrough video is available to aid in navigating the process and can be used in conjunction with the following guide.
Although this guide uses vCenter, NVIDIA AI vWS can be deployed on any NVIDIA vGPU-supported hypervisor. It’s assumed that all vWS requirements, including licensing, are already configured.