Skip to content

NVIDIA/cuda-q-academic

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

744 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

CUDA-Q Academic

🚀 Start Your Journey Here

About

NVIDIA's CUDA-Q Academic is a freely available, open-source collection of interactive educational resources that prepare the next generation of quantum computing professionals by combining high-performance computing with quantum computing. Designed to supplement university quantum computing courses, it enriches classroom instruction and textbook learning with hands-on, interactive modules built using CUDA-Q, NVIDIA's open-source platform for hybrid classical-quantum computing. Developed by NVIDIA in collaboration with universities and tested in real classroom settings, CUDA-Q Academic is organized in modules with topics ranging from a Quick Start to Quantum Computing through Quantum Error Correction, Quantum Algorithm Simulation 101, Dynamics 101, AI for Quantum, Chemistry Simulations, and more. Materials are free to use for educational purposes under Apache-2.0 and CC-BY-NC-4.0; see LICENSE.

Quick Links

Resource Link
Learning Paths (launch modules, build a curriculum) https://nvidia.github.io/cuda-q-academic/learningpath.html
Visualization Gallery (interactive widgets) https://nvidia.github.io/cuda-q-academic/visualization-gallery.html
CUDA-Q Newsletter (updates, tutorials, events) https://www.nvidia.com/en-us/solutions/quantum-computing/cuda-q-newsletter/
Machine-readable curriculum catalog curriculum.json
Guide to CUDA-Q Backends Guide-to-cuda-q-backends.ipynb
Instructor Guide Instructor-Guide.md
Contributing CONTRIBUTING.md
Install CUDA-Q https://nvidia.github.io/cuda-quantum/latest/using/quick_start.html
Coming from another quantum framework?

The CUDA-Q Hello World visualization shows how a quantum circuit diagram translates into CUDA-Q code — a good starting point before you open your first notebook.

Repository Contents

The repository is organized into the learning path modules below. For machine-readable lesson and widget discovery, use curriculum.json. For hosted module overviews, prerequisites, and a curriculum builder, visit the Learning Paths page.

Module Folder Topic
Quick Start to Quantum Computing quick-start-to-quantum/ From zero to a variational algorithm in CUDA-Q
Quantum Algorithm Simulation 101 simulation/ Choosing between state vector, tensor network, MPS, Pauli propagation, and stabilizer simulation
Quantum Information Science Examples qis-examples/ Foundational quantum algorithms to complement QIS courses
Quantum Error Correction 101 qec101/ Classical and quantum codes, decoders, magic-state distillation
Chemistry Simulations chemistry-simulations/ VQE, ADAPT-VQE, QM/MM, Krylov methods, and more.
Quantum Applications for Finance quantum-applications-to-finance/ Quantum walks, portfolio optimization, QChop
QAOA for Max Cut qaoa-for-max-cut/ Divide-and-conquer QAOA with circuit cutting
AI for Quantum ai-for-quantum/ Using AI models to enable quantum computing
Dynamics 101 dynamics101/ GPU-accelerated Schrödinger and Lindblad time evolution
Hybrid Workflows hybrid-workflows/ Hybrid classical–quantum workflow examples

Each module folder contains student notebooks, a module-local README.md, a solutions/ subfolder, and an images/ subfolder with figures.

Instructor Resources

Resource Folder Description
Quantum Computing Group Project quantum-ai-project-template/ Role-based, agentic-AI group project template for deploying a quantum-GPU computing project in a course. See the Group Project section below.
Calibration calibration/ Classroom guide for using NVIDIA's Ising Calibration — an open vision-language model purpose-built for quantum hardware calibration plot analysis. Browser-based NIM playground (no setup, no API key) plus sample plots from the QCalEval dataset. Hands-on tutorial notebooks coming soon.

Quantum Computing Group Project

The CUDA-Q Academic Quantum Computing Group Project is a course-ready template for running a hybrid quantum-GPU computing project as a team assignment. It is designed around agentic AI: each student takes one of four defined roles — Project Lead, Performance Optimization, Quality Assurance, and Technical Marketing — and configures an AI coding agent (Claude Code, Cursor, Windsurf, or any chat-based LLM) from a role-specific starter prompt. Students retain accountability for architecture, classical benchmarking, and honest reporting while delegating implementation to their agents.

Teams (or solo students taking all roles sequentially) move through four milestones — plan, research, build, and showcase — with explicit deliverables at each step. The template ships with:

  • Faculty guides, student-facing templates, and role cards
  • Starter system prompts for each role and the CUDA-Q skill install
  • A deliverable template that captures the plan, benchmark results, and individual retrospectives
  • Suggested project options ranging from a fully scaffolded problem (the MIT iQuHACK 2026 LABS challenge) to student-designed projects

It is not a self-paced module. See quantum-ai-project-template/Teaching-Guide.md for deployment instructions and the folder README for the file inventory.

How to Run

Recommended: launch on NVIDIA Brev. Brev provisions a pre-built CPU or GPU instance with CUDA-Q and all notebook prerequisites already installed.

Launch: https://brev.nvidia.com/launchable/deploy/now?launchableID=env-39dN1v7RucHHgj97LILUlnXjnk5

See brev-instructions.pdf for a step-by-step walkthrough.

Other entry points:

  • qBraid — hosted Jupyter environment with CUDA-Q support. Visit qbraid.com.
  • Amazon Braket — managed quantum development environment that can run CUDA-Q notebooks in Jupyter.
  • Google Colab — each notebook includes a commented-out install cell. Uncomment it, run it to install CUDA-Q and download supporting assets, restart the kernel, and run the notebook.
  • Local — install CUDA-Q directly following the CUDA-Q install guide. Recommended for the largest GPU-accelerated examples.

Each module's README.md contains module-specific run notes.

Contributing

New notebook content follows the schema defined by notebook_template.ipynb: the first markdown cell contains the title plus labeled sections for What You Will Do, Prerequisites, Key Terminology, CUDA-Q Syntax, and a Solutions link. Agents and tooling can rely on this schema to programmatically discover what each notebook covers. AI coding agents working in this repository should read AGENTS.md first.

License & Attribution

CUDA-Q Academic is released under Apache-2.0 and CC-BY-NC-4.0. Developed by NVIDIA in collaboration with university partners; freely available for educational use. This project downloads and installs additional third-party open-source software; review the licenses of those projects before use.

About

This repo contains CUDA-Q Academic materials, including self-paced Jupyter notebook modules for building and optimizing hybrid quantum-classical algorithms using CUDA-Q.

Topics

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors