---
id: nudenet
version: "3.4.2"
license: MIT
license_treatment: permissive
maintenance: active
---

# nudenet — Lightweight Nudity Detection

License: permissive · Maintenance: active · Downloads: 90.5K/mo

## What it is and what it does

NudeNet is a Python library for detecting nudity and exposed body parts in images using ONNX-based neural networks. It wraps ultralytics YOLOv8 models (320n at 320x320 or 640m at 640x640 resolution) to classify image regions into categories like FEMALE_GENITALIA_EXPOSED, FACE_FEMALE, BELLY_EXPOSED, and others. The detector returns bounding boxes with confidence scores for each detected region.

The package is designed for content moderation workflows: you can scan individual images or batches, accept file paths or in-memory image objects (numpy arrays, bytes, or file handles), and optionally blur or censor detected regions. A Docker image is also provided for server-side deployment. The 320n model ships with the package by default; larger models can be downloaded separately.

Use it for:

- Content moderation for user-generated image uploads on social platforms or forums.
- Automated filtering of adult content in image galleries or search results.
- Batch processing of large image datasets to identify and flag potentially sensitive material.
- Client-side nudity detection in browser-based applications using the in-browser demo.
- Building a censoring pipeline to blur or redact exposed regions before display or storage.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Detects nudity in images using lightweight ONNX models, returning bounding boxes and confidence scores for exposed or covered body parts, with optional image censoring.

Yes, if you need lightweight nudity detection. Low install friction, permissive MIT license, active maintenance, and no known vulnerabilities. The 320n model is included by default for quick setup. Trade-off: the maintainer is seeking contributors, so long-term feature development may be limited; verify accuracy on your specific image types before production deployment.

## Install

pip install nudenet
uv add nudenet
poetry add nudenet

## Installing nudenet

Before you install: Low friction: pure Python wheel with three runtime dependencies (numpy, onnxruntime, opencv-python-headless). Actively maintained as of June 2026, with 2427 GitHub stars. Maintainer is seeking contributors but the repo remains current.

License in practice: MIT license is permissive; you can use, modify, and distribute this package freely in commercial or private projects with minimal restrictions.

Quickstart:

pip install nudenet

from nudenet import NudeDetector
detector = NudeDetector()
detections = detector.detect('image.jpg')
print(detections)  # Returns list of dicts with 'class', 'score', 'box'

Requires Python 3.6 or later; onnxruntime and opencv-python-headless must be installed (handled by pip).

Verify before relying:

- Accuracy and false-positive rates across different image types and lighting conditions.
- Performance characteristics (inference time, memory usage) on typical hardware.
- Whether the 320n model is suitable for production use or if 640m is recommended for better accuracy.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 90.5K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

nudity detection in images, NSFW content detection, body part classification, image censoring automation, lightweight object detection, onnx-based image analysis, adult content filtering, content-moderation, computer-vision, onnx-inference

[View on SkillFed](https://skillfed.io/packages/nudenet) · [View on PyPI](https://pypi.org/project/nudenet/)