Repolode
Submit
NVIDIA

NVIDIA/GenerativeAIExamples

A collection of NVIDIA-focused reference workflows for building RAG pipelines, agentic AI and LLM fine-tuning on NIM and NeMo microservices.

View on GitHub
Stars
4,223
Forks
1,102
Open issues
85
Last commit
2 days ago

What is it?

NVIDIA Generative AI Examples is a set of reference workflows showing how to build generative AI systems on NVIDIA's software stack. It covers retrieval-augmented generation (RAG) pipelines, agentic workflows, tool-calling and model fine-tuning, with each example delivered as Jupyter notebooks, sample applications or Docker Compose setups.

The RAG material includes notebooks and end-to-end examples built with LangChain and LlamaIndex connectors, with options for basic, multi-turn, multimodal, structured CSV data and query decomposition. By default the examples call hosted NIM endpoints from the NVIDIA API Catalog, and some can also run on premises. Additional sections cover a Data Flywheel workflow built on NeMo microservices, guardrail and LLM auditing tutorials, and vision NIM workflows for video monitoring and image search.

The repository is aimed at developers who already use LangChain or LlamaIndex and want to plug NVIDIA NIM, NeMo Retriever or related services into their applications.

Key features

  • 01Basic and advanced RAG examples built with LangChain and LlamaIndex
  • 02Tool-calling fine-tuning and evaluation using NeMo microservices
  • 03Agentic RAG notebooks using NeMo Retriever NIM
  • 04Vision workflows for video event monitoring and natural-language image search
  • 05Docker Compose setup for running a sample RAG playground

Use cases

  • Prototyping a RAG chatbot over your own documents using NVIDIA-hosted models
  • Fine-tuning a small LLM for function calling and measuring its accuracy
  • Adding NIM-based retrieval to an existing LangChain or LlamaIndex application
  • Exploring multimodal or structured-data RAG patterns before building a product

Quick start

  1. 1

    Get an NVIDIA API key from the NVIDIA API Catalog

    export NVIDIA_API_KEY=nvapi-...
  2. 2

    Clone the repository

    git clone https://github.com/nvidia/GenerativeAIExamples.git
  3. 3

    Build and run the basic RAG pipeline with Docker Compose

    cd GenerativeAIExamples/RAG/examples/basic_rag/langchain/ && docker compose up -d --build
  4. 4

    Open the sample RAG Playground at https://localhost:8090/ and submit queries

Things to know

  • Examples rely on NVIDIA API Catalog endpoints, which need an NVIDIA API key.
  • Many examples are reference notebooks and workflows rather than a single installable library.

In collections

Open-source AI agent frameworks & appsMicroservice frameworks & service infrastructure

Topics

#rag#llm#nvidia#langchain#llamaindex#agents#fine-tuning#microservices#gpu-acceleration#large-language-models

FAQ

What is GenerativeAIExamples?
A collection of NVIDIA-focused reference workflows for building RAG pipelines, agentic AI and LLM fine-tuning on NIM and NeMo microservices. NVIDIA Generative AI Examples is a set of reference workflows showing how to build generative AI systems on NVIDIA's software stack. It covers retrieval-augmented generation (RAG) pipelines, agentic workflows, tool-calling and model fine-tuning, with each example delivered as Jupyter notebooks, sample applications or Docker Compose setups.
Is GenerativeAIExamples free to use? What license does it have?
GenerativeAIExamples is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is GenerativeAIExamples still maintained?
Actively maintained. The last commit to GenerativeAIExamples was on Nov 6, 2026.
What can I use GenerativeAIExamples for?
Prototyping a RAG chatbot over your own documents using NVIDIA-hosted models Fine-tuning a small LLM for function calling and measuring its accuracy Adding NIM-based retrieval to an existing LangChain or LlamaIndex application Exploring multimodal or structured-data RAG patterns before building a product
How do I install GenerativeAIExamples?
Get an NVIDIA API key from the NVIDIA API Catalog: export NVIDIA_API_KEY=nvapi-... Clone the repository: git clone https://github.com/nvidia/GenerativeAIExamples.git Build and run the basic RAG pipeline with Docker Compose: cd GenerativeAIExamples/RAG/examples/basic_rag/langchain/ && docker compose up -d --build Open the sample RAG Playground at https://localhost:8090/ and submit queries
Where can I get GenerativeAIExamples?
The source code and installation instructions are on GitHub: https://github.com/NVIDIA/GenerativeAIExamples

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects