New NVIDIA AI Blueprints for Building Agentic AI Applications
Introduction
NVIDIA has released new AI Blueprints for building agentic AI applications, enabling developers to create custom AI agents that can reason, plan, and take action to quickly analyze large amounts of data, summarize and distill real-time insights from video, PDF, and other images.
Agentic AI Orchestration Tools
Agentic AI represents the next wave in the evolution of generative AI, enabling applications to move beyond simple chatbot interactions to tackle complex, multi-step problems through sophisticated reasoning and planning. The agentic AI orchestration layer from NVIDIA partners provides the glue needed for AI agents to effectively work together.
New Partner Blueprints
The new partner blueprints, available from agentic AI orchestration leaders, offer integrations with NVIDIA AI Enterprise software, including NIM microservices and NVIDIA NeMo Retriever, to boost retrieval accuracy and reduce latency of agent workflows. The following blueprints are now available:
- CrewAI: Code documentation for software development using Llama 3.3 70B NIM microservices and the NVIDIA NeMo Retriever embedding NIM microservice.
- Daily: Voice agent blueprint using the company’s open-source Pipecat framework, NVIDIA Riva automatic speech recognition and text-to-speech NIM microservice, and Llama 3.3 70B NIM microservice for real-time conversational AI.
- LangChain: Structured report generation blueprint using Llama 3.3 70B NVIDIA NIM microservices and the LangGraph framework.
- LlamaIndex: Document research assistant for blog creation blueprint harnessing NVIDIA NIM microservices and NeMo Retriever for automatic research, outlining, and generation of compelling content with source attribution.
- Weights & Biases: AI virtual assistant blueprint featuring W&B Weave capability, Llama 1.1 70B NIM microservice, and NVIDIA NeMo Retriever for debugging, evaluating, and iterating AI applications.
New Blueprints from NVIDIA
NVIDIA is also introducing the following new blueprints:
- PDF to Podcast: A recipe for developers to turn multiple long and complex PDFs into AI-generated readouts, extracting images, tables, and text, and converting the data into easily digestible audio content while keeping data secure.
- Video Search and Summarization: A blueprint for building AI agents that can analyze and summarize video content, extracting key points and generating a concise summary.
NVIDIA Omniverse Blueprints
NVIDIA is also launching four additional Omniverse Blueprints that make it easier for developers to build simulation-ready digital twins for physical AI. These blueprints are designed for industries such as automotive, technology, manufacturing, and consumer goods.
Conclusion
These new AI Blueprints from NVIDIA and its partners empower developers to create advanced AI applications that can transform industries. With the introduction of agentic AI, developers can build AI agents that can reason, plan, and take action to quickly analyze large amounts of data, summarize and distill real-time insights from video, PDF, and other images.
FAQs
Q: What are agentic AI blueprints?
A: Agentic AI blueprints are a new category of partner blueprints for building agentic AI applications.
Q: What are the benefits of agentic AI blueprints?
A: Agentic AI blueprints enable developers to create custom AI agents that can reason, plan, and take action to quickly analyze large amounts of data, summarize and distill real-time insights from video, PDF, and other images.
Q: What are the new partner blueprints available?
A: The new partner blueprints are available from agentic AI orchestration leaders, including CrewAI, Daily, LangChain, LlamaIndex, and Weights & Biases.
Q: How can I get started with AI Blueprints?
A: You can get started with AI Blueprints by visiting the NVIDIA website and exploring the various blueprints available.
