# cornellius-gp/gpytorch

A highly efficient implementation of Gaussian Processes in PyTorch

Repository: https://github.com/cornellius-gp/gpytorch
Canonical: https://ross.abutalabs.com/products/gpytorch
Language: Python
License: MIT
License Family: permissive
Topics: pytorch, gaussian-processes, gpu-acceleration
Last push: 2026-07-10T22:03:14+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 91, release rhythm 60, longevity 100
- inputs: {"age_days": 3372, "days_push": 54, "days_rel": 187, "gap_med": 44, "n_releases_24m": 8}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3906, forks 598 (observed 2026-08-28T04:08:28.710043+00:00)

## What it is
GPyTorch is a Gaussian process library built on PyTorch, designed for scalable, flexible, and modular GP models. It uses linear algebra techniques like preconditioned conjugate gradients for efficient inference and strong GPU acceleration.

## Use cases
- fit gaussian process regression models in pytorch
- scale gaussian processes to large datasets on gpu
- build deep kernel learning models
- implement scalable GP methods like SKI/KISS-GP and stochastic variational inference
- combine gaussian processes with neural networks
- quantify uncertainty in model predictions

## When to choose
- you need Gaussian process models with GPU acceleration
- you want to integrate GPs with deep learning frameworks
- you need state-of-the-art scalable GP algorithms
- your project already uses PyTorch

## When to avoid
- you need a simple off-the-shelf GP without PyTorch dependency
- you work outside the Python/PyTorch ecosystem
- you need classical ML models beyond Gaussian processes

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, math, gpu-computing
- domain: machine-learning, data-science
- platform: python, cross-platform
- tags: gaussian-processes, pytorch, bayesian-inference, probabilistic-modeling, kernel-methods, algorithms, gpu

## Member repositories
- cornellius-gp/gpytorch (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:28.710043+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 2026-08-29T18:25:06.847803+00:00, confidence not recorded.
  - readme: https://github.com/cornellius-gp/gpytorch (fetched 2026-08-28T04:08:28.710043+00:00, sha b5c3e3b04cff)
  - registry_pypi: https://pypi.org/pypi/gpytorch/json (fetched 2026-08-29T09:19:09.902959+00:00, sha ade97b9db662)
- Data as of 2026-08-30T08:39:29.467469+00:00.