# Xtra-Computing/thundersvm

ThunderSVM: A Fast SVM Library on GPUs and CPUs

Repository: https://github.com/Xtra-Computing/thundersvm
Canonical: https://ross.abutalabs.com/products/thundersvm
Language: C++
License: Apache-2.0
License Family: permissive
Topics: classification, regression, one-class-learning, cuda, c-plus-plus, gpu, libsvm
Last push: 2024-04-01T08:11:14+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 4283, "days_push": 884, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1621, forks 222 (observed 2026-08-28T04:05:12.446615+00:00)

## What it is
ThunderSVM is a fast SVM library that accelerates support vector machine training on GPUs (via CUDA) and multi-core CPUs. It supports all LibSVM functionalities (SVC, SVR, one-class SVM, probabilistic SVMs) with LibSVM-compatible command-line options and Python, R, Matlab, and Ruby interfaces.

## Use cases
- train SVM classifiers on large datasets with GPU acceleration
- speed up libsvm workloads without changing command-line options
- fit support vector regression models faster than libsvm
- use SVMs from Python with a scikit-learn-like interface
- run one-class SVM anomaly detection on GPU
- train probabilistic SVMs quickly
- use SVMs on machines without GPUs via multi-core CPU support

## When to choose
- you need SVM training that is much faster than libsvm on large datasets
- you have an NVIDIA GPU with CUDA and want drop-in LibSVM-compatible behavior
- you want SVM access from Python, R, Matlab, or Ruby

## When to avoid
- you need deep learning or gradient boosting instead of SVMs (consider ThunderGBM for GBDT)
- you have no GPU and a small dataset where libsvm suffices
- you need actively developed cutting-edge features or frequent releases

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning
- domain: machine-learning, data-science
- platform: windows, python, cpp
- tags: svm, libsvm-compatible, cuda, gpu-acceleration, scikit-learn-interface, classification, regression, one-class-svm, linux, macos, gpu

## Member repositories
- Xtra-Computing/thundersvm (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:12.446615+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 2026-08-30T03:49:05.865726+00:00, confidence not recorded.
  - readme: https://github.com/Xtra-Computing/thundersvm (fetched 2026-08-28T04:05:12.446615+00:00, sha 1568d63becbe)
- Data as of 2026-08-30T08:39:29.467469+00:00.