# open-edge-platform/anomalib

An anomaly detection library comprising state-of-the-art algorithms and features such as experiment management, hyper-parameter optimization, and edge inference.

Repository: https://github.com/open-edge-platform/anomalib
Canonical: https://ross.abutalabs.com/products/anomalib
Homepage: https://anomalib.readthedocs.io/en/latest/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: anomaly-localization, anomaly-segmentation, anomaly-detection, unsupervised-learning, neural-network-compression, openvino, geti
Last push: 2026-08-26T13:39:00+00:00

## Health v2 (maintenance only)
Score: 98/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 95, longevity 100
- inputs: {"age_days": 1765, "days_push": 7, "days_rel": 39, "gap_med": 13, "n_releases_24m": 16}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 6088, forks 974 (observed 2026-08-28T04:09:34.932692+00:00)

## What it is
Anomalib is a Python deep learning library for anomaly detection, offering state-of-the-art unsupervised algorithms for detecting and localizing anomalies in images. It includes experiment management, hyper-parameter optimization, and export/edge inference support via OpenVINO.

## Use cases
- detect defects in industrial images
- train an unsupervised anomaly detection model
- segment anomalous regions in images
- benchmark anomaly detection algorithms
- deploy anomaly detection to edge devices
- optimize hyperparameters for anomaly models
- visual inspection of manufactured parts

## When to choose
- you need ready-to-use state-of-the-art anomaly detection or localization models
- you want PyTorch-based training with experiment tracking and HPO built in
- you need to export models for edge inference with OpenVINO

## When to avoid
- you need supervised classification rather than unsupervised anomaly detection
- you need anomaly detection on time series or tabular data rather than images
- you want a lightweight inference-only tool without training features

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, computer-vision, image-processing, benchmarking, llm-training
- domain: computer-vision, machine-learning, deep-learning, image-processing
- platform: python, cross-platform
- tags: anomaly-detection, anomaly-segmentation, unsupervised-learning, openvino, pytorch, edge-inference, industrial-inspection, gpu

## Member repositories
- open-edge-platform/anomalib (main) score 98

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:34.932692+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 2026-08-29T17:48:18.225758+00:00, confidence not recorded.
  - readme: https://github.com/open-edge-platform/anomalib (fetched 2026-08-28T04:09:34.932692+00:00, sha 0701dc0da9e2)
- Data as of 2026-08-30T08:39:29.467469+00:00.