NVIDIA Accelerated Computing Hub
This repository is a home for open learning materials related to GPU computing. You will find user guides, tutorials, and other works freely available for all learners interested in GPU computing.
The intention is to create a living project where documents, examples, best practices, optimizations, and new features can become both visible and accessible to users quickly and easily.
Tutorials and Syllabi
The following interactive tutorials are available and can be used on NVIDIA Brev or Google Colab. See the documentation for more information on creating and deploying Brev Launchables using this content.
| Content | Docker Compose | Brev Instance | Brev Provider |
|---|---|---|---|
| CUDA Tile Tutorial | Generated Compose | A10G | AWS or any provider with Flexible Ports |
| CUDA C++ Tutorial | docker-compose.yml | L40S, L4, or T4 | Crusoe or any other with Flexible Ports |
| Standard Parallelism Tutorial | docker-compose.yml | 4xL4, 2xL4, 2xL40S, or 1x L40S | GCP, AWS, or any other with Flexible Ports and Linux 6.1.24+, 6.2.11+, or 6.3+ (for HMM) |
| Accelerated Python Tutorial | docker-compose.yml | L40S, L4, or T4; A10G or newer Ampere/Ada/Blackwell for CUDA Tile; 4xL4 or 2xL4 for distributed | Crusoe or any other with Flexible Ports; host driver must support CUDA 13 |
| NVIDIA Warp Tutorial | docker-compose.yml | L40S, L4, or T4 | Crusoe or any other with Flexible Ports |
| Sim2Real Blogs | docker-compose.yml | L40S, L4, or T4 | Crusoe or any other with Flexible Ports |
| nvmath-python Tutorial | docker-compose.yml | 4xL4, 2xL4, 2xL40S, or 1x L40S | Crusoe or any other with Flexible Ports |
| CUDA Python Tutorial - CuPy, cuDF, CCCL, & Kernels - 8 Hours | docker-compose.yml | L40S, L4, or T4 | Crusoe or any other with Flexible Ports |
| CUDA Python Tutorial - CuPy, SIMT, & Tile - 8 Hours | docker-compose.yml | A10G | AWS or any provider with Flexible Ports |
| CUDA Python Tutorial - cuda.core & CCCL - 2 Hours | docker-compose.yml | L40S, L4, or T4 | Crusoe or any other with Flexible Ports |
| PyHPC Tutorial - NumPy, CuPy, & mpi4py - 4 Hours | docker-compose.yml | 4xL4, 2xL4, 2xL40S, or 1x L40S | Crusoe or any other with Flexible Ports; host driver must support CUDA 13 |
| PyHPC Tutorial - CuPy, Kernels, MPI, JAX, OMP, Interop - 2 Days | docker-compose.yml | L40S, L4, or T4 | Crusoe or any other with Flexible Ports; host driver must support CUDA 13 |
| GPU Deployment Tutorial | |||
| Newton Robot Tasks Tutorial | docker-compose.yml | NVIDIA GPU with a CUDA 13.1-compatible driver; local CPU path available | Any provider with Flexible Ports; see tutorial requirements |
License
All written materials (user guides, documentation, presentations) are subject to Creative Commons CC BY-NC-SA 4.0.
All codes (notebook code, coding examples) are subject to Apache License, Version 2.0.
Contact
For additional help, please use NVIDIA's CUDA Developer Forums.