Solution overview

Validated full-stack solution focused on AI Enterprise solutions. The validated environment provides a highly performant environment to test the deployment and tuning of different NVIDIA NIMs and Blueprints

Powered by NVIDIA L40s GPUs, NVIDIA Microservices, and Red Hat OpenShift Kubernetes platform. The environment provides the perfect platform for customers to evaluate NIMs such as PDF Ingestion, NeMo RAG, and Optical Character Recognition.

FlashBlade//S ensures high-speed, reliable data access for AI workloads, from training to inference.

Advanced AI Use Cases

  • PDF Ingestion (NV-Ingest)
    • A scalable microservice for extracting text, tables, charts, and images from documents, including PDFs, Word, and PowerPoint files.
  • RAG (Retrieval-Augmented Generation)
    • Enables AI applications to access and utilize data from various sources through a retriever model.
  • OCR (Optical Character Recognition)
    • A powerful NIM for extracting text from images with high accuracy.
  • Enterprise RAG Pipeline
    • Automates the process of extracting, embedding, and indexing multimodal data for semantic search.

Pre-Deployed NVIDIA AI Blueprint

The NVIDIA AI Blueprint: Video Search and Summarization

Description

Insightful, accurate, and interactive video analytics AI agents enable a range of industries to make better decisions faster. These AI agents are given tasks through natural language and can perform complex operations like video summarization and visual question-answering, unlocking entirely new application possibilities. The NVIDIA AI Blueprint makes it easy to get started building and customizing video analytics AI agents for video search and summarization — all powered by generative AI, vision language models (VLMs) like Cosmos Nemotron VLMs, large language models (LLMs) like Llama Nemotron LLMs, NVIDIA NeMo Retriever, and NVIDIA NIM.

Launch the lab to see it in action.  Chose a sample video file and click the summarization button to see the results.   

Technologies