# NVIDIA/CUDALibrarySamples

CUDA Library Samples

Repository: https://github.com/NVIDIA/CUDALibrarySamples
Canonical: https://ross.abutalabs.com/products/cudalibrarysamples
Language: Cuda
License: Apache-2.0
License Family: permissive
Topics: cufft, curand, cusolver, cusparse, nvjpeg, cudss, cutenros, mathdx, nppcublas, nvcomp, nvjpeg2000, nvtiff, cuda, gpu, linear-algebra
Last push: 2026-08-24T17:42:56+00:00

## Health v2 (maintenance only)
Score: 77/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 35, longevity 100
- inputs: {"age_days": 2673, "days_push": 9, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2489, forks 471 (observed 2026-08-28T04:06:55.684696+00:00)

## What it is
NVIDIA's official collection of sample programs demonstrating the use of GPU-accelerated CUDA libraries such as cuBLAS, cuFFT, cuSOLVER, cuSPARSE, cuRAND, NPP, nvJPEG, and nvCOMP. Each sample provides practical, runnable C++/CUDA code showing how to apply these libraries to math, linear algebra, signal and image processing, and compression workloads.

## Use cases
- learn how to use cuBLAS for GPU linear algebra
- example code for FFT on GPU with cuFFT
- JPEG encode decode on GPU with nvJPEG
- sparse matrix solver examples with cuSPARSE and cuDSS
- GPU data compression with nvCOMP
- random number generation on GPU with cuRAND
- image and video processing samples with NPP

## When to choose
- you are developing CUDA applications and want official, working examples of NVIDIA's accelerated libraries
- you need reference code for GPU linear algebra, FFT, solvers, or image codecs
- you want to evaluate which CUDA library fits your HPC workload

## When to avoid
- you need a production-ready library rather than example code
- you have no NVIDIA GPU or CUDA toolchain
- you want high-level frameworks or Python APIs instead of C++/CUDA samples

## Facets
- artifact type: learning-resource
- maturity: active
- function: image-processing, compression, math, graphics, sdk
- domain: gpu-computing, developer-tools, image-processing, tutorials
- platform: cpp, windows
- tags: cuda, cublas, cufft, cusolver, cusparse, curand, npp, nvjpeg, nvcomp, linear-algebra, sample-code, nvidia, algorithms, gpu, linux

## Member repositories
- NVIDIA/CUDALibrarySamples (main) score 77

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:55.684696+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:27:50.125825+00:00, confidence not recorded.
  - readme: https://github.com/NVIDIA/CUDALibrarySamples (fetched 2026-08-28T04:06:55.684696+00:00, sha d894b17eff82)
- Data as of 2026-08-30T08:39:29.467469+00:00.