NVIDIA NIM Agent Blueprints Fast-Forward the Next Wave of Enterprise Generative AI
NVIDIA NIM Agent Blueprints are pretrained AI workflows tailored for specific use cases. They can include sample applications built with NVIDIA NIM and partner microservices, one or more AI agents, reference code, customization documentation and a Helm chart for deployment.
With NVIDIA NIM Agent Blueprints, developers can gain a head start on creating their own applications using NVIDIA's advanced AI tools and end-to-end development experience for each use case. The blueprints are designed to be modified and enhanced, and allow developers to leverage both information retrieval and agent-based workflows capable of performing complex tasks. The first NVIDIA NIM Agent Blueprints available are:
Digital Humans for Customer Service
This digital human for customer service NIM Agent Blueprint is powered by NVIDIA Tokkio to bring enterprise applications to life with a 3D animated digital human interface. With an approachable, human-like interface, customer service applications can provide better user experiences with faster resolutions than generative AI powered applications with just a text or speech interface.
This digital humans for virtual screening NIM Agent Blueprint is designed to integrate within your existing generative AI applications built using retrieval-augmented generation (RAG). Use this blueprint to start evolving your applications running in your data center, in the cloud, or at the edge, to include a full digital human interface.
Multimodal PDF Data Extraction for Enterprise RAG
The multimodal PDF data extraction for enterprise RAG NIM Blueprint uses NVIDIA NeMo Retriever NIM microservices to unlock insights from massive volumes of enterprise data. With this enterprise-scale multimodal document retrieval blueprint, developers can create digital humans, AI agents, or customer service chatbots that can quickly gain expertise on topics captured within their corpus of data.
The multimodal PDF data extraction blueprint is designed to enhance generative AI applications with RAG capabilities which can be connected to proprietary data–wherever it resides. Use this workflow to supercharge your RAG applications with unprecedented intelligence.
Generative Virtual Screening for Drug Discovery
The generative virtual screening for drug discovery NIM Agent Blueprint is powered by three NVIDIA NIM microservices that address core tasks in drug discovery:
- First, the target protein sequence is passed to the AlphaFold2 NIM, which accurately determines that protein's structure.
- Second, 10-100 initial chemical structures are passed to the MolMIM NIM, which is used to generate diverse small molecules for exploring that chemical space.
- Third, those initial structures are scored and ranked for multiple characteristics of drugs, such as solubility, Quantitative Estimate of Drug-likeness (QED), and positive feedback from human chemists.
- Fourth, those scores inform the MolMIM NIM during iterative generation cycles to rank the generative molecules until they reach a desirable threshold for further testing. The MolMIM NIM can optimize for all these features simultaneously, enabling up to 80x enrichments of optimized molecules in its output relative to brute force methods.
- Fifth, the generated molecules are filtered by their ability to bind the input target protein structure with the DiffDock NIM. The DiffDock NIM can predict protein-ligand poses up to 7.6x faster than the baseline model.
- Finally, optimized molecules are returned to the user for further lab testing.
WWT experts are ready to leverage the capabilities of the AI Proving Ground and the Advanced Technology Center (ATC) to support the latest in NVIDIA NIM Agent Blueprints.
Connect with our NVIDIA Experts
Learn more about NVIDIA NIM Agent Blueprints and WWT
NVIDIA
World Wide Technology Expands AI Proving Ground with NVIDIA NIM Agent Blueprints
WWT's AI Proving Ground Now Equipped with NVIDIA NIM Agent Blueprints
WWT's AI Proving Ground with NVIDIA NIM Agent Blueprints: Force Multipliers for Scaling Enterprise AI