# anderskm/gputil

A Python module for getting the GPU status from NVIDA GPUs using nvidia-smi programmically in Python

Repository: https://github.com/anderskm/gputil
Canonical: https://ross.abutalabs.com/products/gputil
Language: Python
License: MIT
License Family: permissive
Last push: 2026-07-18T05:34:16+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 93, release rhythm 8, longevity 100
- inputs: {"age_days": 3516, "days_push": 46, "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 1214, forks 131 (observed 2026-08-28T04:04:00.901579+00:00)

## What it is
GPUtil is a Python module that queries NVIDIA GPU status (memory usage, load, availability) by wrapping the nvidia-smi command-line tool. It returns an ordered list of available GPUs, primarily intended for GPU selection in deep learning workflows.

## Use cases
- find a free nvidia gpu before launching a training job
- monitor gpu memory and load from python
- select the least utilized gpu in tensorflow or pytorch
- check gpu utilization programmatically
- restrict a script to a single available gpu
- monitor gpu usage in a background thread

## When to choose
- you need a lightweight, dependency-free way to query NVIDIA GPU status in Python
- you want to automatically pick an available GPU for deep learning workloads
- you need Python 2 and 3 compatibility for GPU discovery

## When to avoid
- you need AMD, Intel, or Apple Silicon GPU monitoring
- you want fine-grained metrics or process-level GPU data beyond what nvidia-smi exposes
- you need an actively developed library with modern feature support

## Facets
- artifact type: library
- maturity: maintenance
- function: monitoring, gpu-computing, machine-learning
- domain: gpu-computing, machine-learning, developer-tools
- platform: python, windows
- tags: nvidia, nvidia-smi, gpu-status, gpu-selection, cuda, system-monitoring, linux, macos

## Member repositories
- anderskm/gputil (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:00.901579+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 2026-08-30T06:17:20.934099+00:00, confidence not recorded.
  - readme: https://github.com/anderskm/gputil (fetched 2026-08-28T04:04:00.901579+00:00, sha fcdfed837f45)
  - registry_pypi: https://pypi.org/pypi/gputil/json (fetched 2026-08-29T12:25:30.908583+00:00, sha b88ca4c313e2)
- Data as of 2026-08-30T08:39:29.467469+00:00.