Skip to content
SPECIFICTION
Get listed

Anomalib

github.com

Free plan Open source WebWindowsLinuxAPISelf-hosted
Visit github.com
Spec score 9.1 #1 of 22 in Anomaly Detection Software

What it is

Anomalib is a free, Apache-2.0-licensed deep learning library for developing, benchmarking and deploying anomaly detection algorithms, especially for images and videos. Its modular Python API and command-line interface support training, inference and benchmarking, and the project includes ready-to-use algorithms and benchmark datasets. Model implementations are based on Lightning; inference options include Torch, Lightning, Gradio and OpenVINO. Most models can be exported to OpenVINO Intermediate Representation for accelerated inference on Intel hardware. Installation choices include CPU, NVIDIA CUDA on Linux or Windows, AMD ROCm on Linux and Intel XPU on Linux. Documented experiment-tracking integrations include Weights & Biases, Comet.ml and TensorBoard through PyTorch Lightning loggers. Anomalib Studio adds a low/no-code web application that accepts USB or IP cameras or image folders and can send output to industrial pipelines through ROS messages or MQTT. Studio is offered as a Docker container or standalone application, but is a pre-release under active development; features may change and functionality may be incomplete or unstable. Intel GPU training currently supports one GPU.

Spec sheet

What it does
Anomalib is a deep-learning library for benchmarking, developing and deploying visual anomaly-detection algorithms on image and video data.Read 15 Oct 2026 · github.com
Purpose
Anomalib is a deep learning library for benchmarking, developing, and deploying anomaly detection algorithms, with a focus on detecting or localizing anomalies in images and videos.Read 4 Oct 2026 · github.com
Deployment
Inference options include Torch, Lightning, Gradio, and OpenVINO.Read 4 Oct 2026 · github.com
Security
The project documents continuous security scanning with CodeQL, Semgrep, Bandit, Zizmor, Trivy, and Dependabot, and directs vulnerability reports to Intel's vulnerability handling guidelines.Read 4 Oct 2026 · github.com
License
The repository identifies its license as Apache-2.0.Read 4 Oct 2026 · github.com
Algorithm collection
The documentation lists ready-made image models including PatchCore, PaDiM, DRAEM, FastFlow, GANomaly, Reverse Distillation, STFPM and WinCLIP.Read 15 Oct 2026 · anomalib.readthedocs.io
Algorithms and datasets
The project describes its collection as ready-to-use deep learning anomaly detection algorithms and benchmark datasets.Read 4 Oct 2026 · github.com
API and CLI
It provides a modular Python API and command-line interface for training, inference, benchmarking and hyperparameter optimization.Read 15 Oct 2026 · anomalib.readthedocs.io
Breaking changes
The 2.7.0 changelog removes the legacy ONNX exporter, several deprecated APIs and older checkpoint migration support.Read 15 Oct 2026 · github.com
Core dependency requirement
The v2.7.0 README badge states minimum PyTorch 2.10+, Lightning 2.2+ and OpenVINO 2024.0+.Read 15 Oct 2026 · github.com
Datasets
Anomalib includes datamodules for image, video and depth datasets and documents public benchmark datasets.Read 15 Oct 2026 · anomalib.readthedocs.io
Detection scope
The project targets anomaly detection, anomaly localization and anomaly segmentation for images and videos.Read 15 Oct 2026 · github.com
Documentation depth
The documentation includes reference APIs, CLI, data and model guides, inference, metrics, pipelines, tutorials and developer guides.Read 15 Oct 2026 · anomalib.readthedocs.io
Edge inference
Most models can be exported to OpenVINO Intermediate Representation for accelerated inference on Intel hardware.Read 4 Oct 2026 · github.com
Experiment tracking
Optional logging integrations include Weights & Biases, Comet.ml, TensorBoard and MLflow.Read 15 Oct 2026 · github.com
Hardware extras
Installation extras cover CPU, NVIDIA CUDA 12.6 or 13.0, AMD ROCm and Intel XPU backends.Read 15 Oct 2026 · github.com
Hardware support
Installation options include CPU, CUDA on Linux or Windows with NVIDIA GPUs, ROCm on Linux with AMD GPUs, and Intel XPU on Linux.Read 4 Oct 2026 · github.com
Independent distribution evidence
SourceForge hosts an unaffiliated mirror and lists zero downloads for the displayed release folders.Read 15 Oct 2026 · sourceforge.net
Independent package status
PyPI classifies Anomalib 2.7.0 as production/stable software for developers and scientific or engineering use.Read 15 Oct 2026 · pypi.org
Independent research description
The peer-reviewed project paper describes Anomalib as an open-source, plug-and-play library with experiment trackers, visualizers, hyperparameter optimizers and OpenVINO optimization.Read 15 Oct 2026 · arxiv.org
Industrial outputs
Studio can send outputs into industrial pipelines through ROS messages and MQTT.Read 15 Oct 2026 · github.com
Inference options
Inference support covers Torch, Lightning, Gradio and OpenVINO workflows.Read 15 Oct 2026 · github.com
Installation
The recommended standard installation is through PyPI using pip or uv, with source installation available for development.Read 15 Oct 2026 · github.com
Intel GPU limit
Training on Intel GPUs currently supports only a single GPU and was tested on Arc 750 and Arc 770.Read 15 Oct 2026 · github.com
Licence
The repository and package metadata identify Apache License 2.0 as the project licence.Read 15 Oct 2026 · github.com
Maker founding
Intel was founded by Robert Noyce and Gordon Moore in 1968.Read 15 Oct 2026 · intel.com
Maker headquarters
Intel Corporation’s principal executive office is at 2200 Mission College Boulevard, Santa Clara, California.Read 15 Oct 2026 · sec.gov
Maker legal name
The package is published from the open-edge-platform/anomalib repository maintained under Intel’s project security policy; Intel’s legal company name is Intel Corporation.Read 15 Oct 2026 · sec.gov
Maker platform
Intel announced its modular open Edge Platform for developing, deploying, securing and managing edge AI applications on February 25, 2024.Read 15 Oct 2026 · intel.com
Maker size
Intel reported 85,100 employees as of December 27, 2025.Read 15 Oct 2026 · sec.gov
Model framework
Its model implementations are built on PyTorch Lightning to reduce training boilerplate.Read 15 Oct 2026 · github.com
OpenVINO export
Most models can be exported to OpenVINO Intermediate Representation for accelerated Intel-hardware inference.Read 15 Oct 2026 · github.com
Python requirement
The current package metadata supports Python 3.10, 3.11 and 3.12.Read 15 Oct 2026 · pypi.org
Rating on SourceForge
Not rated out of 5 from 0 reviews; the page shows an empty user-ratings area and no written reviews.Read 15 Oct 2026 · sourceforge.net
Release news
Version 2.7.0 was published on October 6, 2026 on PyPI.Read 15 Oct 2026 · pypi.org
Security reporting
Vulnerability reports are directed to Intel’s vulnerability-handling guidelines.Read 15 Oct 2026 · github.com
Security scanning
The project documents continuous CodeQL, Semgrep, Bandit, Zizmor, Trivy and Dependabot checks.Read 15 Oct 2026 · github.com
Self-hosting
The library and Studio can be run locally from source, with documented backend and frontend setup commands.Read 15 Oct 2026 · github.com
Studio
Anomalib Studio is a low/no-code web application that accepts USB or IP cameras or image folders and can output to industrial pipelines through ROS messages or MQTT.Read 4 Oct 2026 · github.com
Studio application
Anomalib Studio is a low/no-code web application for training and deploying models from USB cameras, IP cameras or image folders.Read 15 Oct 2026 · github.com
Studio availability
Studio is described as a pre-release under active development, with features that may change and functionality that may be incomplete or unstable; it is offered as a Docker container or standalone application.Read 4 Oct 2026 · github.com
Studio distribution
Studio is distributed as a Docker container or standalone application.Read 15 Oct 2026 · github.com
Studio maturity
The project labels Studio pre-release and warns that functionality may be incomplete, unstable or subject to change.Read 15 Oct 2026 · github.com
Training and benchmarking
It provides a modular Python API and CLI for training, inference, and benchmarking.Read 4 Oct 2026 · github.com
Video models
The reference documentation includes an AI-VAD video anomaly-detection model.Read 15 Oct 2026 · anomalib.readthedocs.io

Plans and prices

Open-source Anomalib library Free Apache-2.0 licensed · installable from PyPI or source · self-hosted Read 15 Oct 2026 · github.com
Open-source library Free Apache-2.0 licensed library · Install from PyPI or source Read 4 Oct 2026 · github.com

Every Anomalib plan and price →

Strong points and limits

Strong points

  • Free and Apache-2.0 licensed.
  • Includes algorithms and benchmark datasets.
  • Supports several inference options.
  • Can export most models for Intel inference.
  • Studio accepts cameras or image folders.

Limits

  • Studio may be incomplete or unstable.
  • Intel GPU training supports only one GPU.
  • AMD ROCm installation is Linux-only.
  • Intel XPU installation is Linux-only.

Who it suits

Anomalib suits developers working on image or video anomaly detection who want Python and CLI tooling for training, inference and benchmarking. Studio may interest teams seeking a low/no-code interface, provided they account for its pre-release status.

Verdict

Anomalib provides a broad toolkit for developing and deploying image and video anomaly detection, with multiple hardware and inference paths. Treat Studio as a pre-release, and note the single-GPU limit for Intel GPU training.

Where Specifiction places it

Alternatives to AnomalibAll alternatives →