NVIDIA launches NIM Agent Blueprints for scalable AI solutions

NVIDIA has announced the release of NVIDIA NIM Agent Blueprints. The release comprises of a comprehensive catalogue of pre-trained, customisable AI workflows designed to enable enterprise developers to build and deploy generative AI applications for common use cases.

Such use cases include customer service, drug discovery, and data extraction from PDFs. The workflows are part of the NVIDIA AI Enterprise platform.

According to NVIDIA, NIM Agent Blueprints provide developers with a starting point for creating AI applications that utilise one or more AI agents. The blueprints include sample applications built with NVIDIA NeMo, NVIDIA NIM, and partner microservices. They also come with reference code, customisation documentation, and deployment charts.

With these tools, enterprises can modify the blueprints using their own business data and run generative AI applications across various data centres and cloud environments. The first set of blueprints available includes digital human workflows for customer service, generative virtual screening workflows for computer-aided drug discovery, and multimodal PDF data extraction workflows.

"Generative AI is advancing at lightspeed. Frontier model capabilities are growing exponentially with a continuous stream of new applications," said Jensen Huang, founder and CEO of NVIDIA. "The enterprise AI wave is here. With the NVIDIA AI Enterprise toolkit - including NeMo, NIM microservices, and the latest NIM Agent Blueprints - our expansive partner ecosystem is poised to help enterprises customise open-source models, build bespoke AI applications, and deploy them seamlessly across any cloud, on-premises or at the edge."

NVIDIA's global system integration and technology solutions partners, such as Accenture, Deloitte, SoftServe, World Wide Technology (WWT), Cisco, Dell Technologies, Hewlett Packard Enterprise (HPE), and Lenovo, are playing a critical role in bringing these blueprints to enterprises worldwide. These partners offer full-stack NVIDIA-accelerated infrastructures and solutions to expedite NIM Agent Blueprint deployments.

The digital human NIM Agent Blueprint for customer service, for example, is designed to help enterprises create engaging user experiences with a 3D animated avatar interface. Gartner has predicted that 80% of conversational offerings will incorporate generative AI by 2025, up from 20% in 2024. This workflow leverages NVIDIA Tokkio technologies alongside other NVIDIA software to integrate smoothly with existing enterprise generative AI applications.

The multimodal PDF data extraction workflow is another significant addition, enabling enterprises to unlock insights from large volumes of PDF data. This workflow uses NVIDIA NeMo Retriever NIM microservices and allows developers to build high-accuracy retrieval pipelines capable of being deployed wherever enterprise data resides.

The generative virtual screening blueprint aims to speed up drug discovery by facilitating the identification and optimisation of drug-like molecules. This blueprint incorporates NVIDIA NIM microservices like AlphaFold2, MolMIM, and DiffDock to accelerate molecular screening processes.

NVIDIA plans to release additional blueprints monthly, covering workflows for customer experience, content generation, software engineering, and product research and development.

Accenture, one of the significant partners, will add NVIDIA NIM Agent Blueprints to its Accenture AI Refinery. "Across industries, generative AI is acting as a catalyst for companies looking to reinvent with tech, data and AI," said Julie Sweet, chair and CEO of Accenture.

Global consulting firm Deloitte also plans to incorporate these blueprints into its portfolio. "By embedding NVIDIA's NIM Agent Blueprints into enterprise solutions built on NVIDIA NIM microservices, Deloitte is engaging with our clients to innovate faster, unlock new growth opportunities and define AI-competitive advantage," said Jason Girzadas, CEO of Deloitte US.

SoftServe and WWT are among other partners readying to integrate these blueprints into their generative AI solutions portfolios, aiming to facilitate the adoption and deployment of tailored AI applications across various enterprises.

Enterprises will be able to develop and deploy NIM Agent Blueprints on NVIDIA AI platforms using compute, networking, and software provided by NVIDIA's global server manufacturing partners. This collaboration includes infrastructure solutions like Cisco Nexus HyperFabric AI clusters, the Dell AI Factory with NVIDIA, and NVIDIA AI Computing by HPE.

These partnerships affirm NVIDIA's commitment to facilitating the enterprise adoption of generative AI, providing a robust ecosystem for rapid AI application development and deployment.