cornellius-gp/gpytorch

Health v2 · maintenance only

Adoption not part of the score

3906 stars · 598 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

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

library · maturity stable

machine-learning math gpu-computing machine-learning data-science python cross-platform gaussian-processes pytorch bayesian-inference probabilistic-modeling kernel-methods algorithms gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="cornellius-gp/gpytorch")