Anomalib is a deep learning library for developing, benchmarking, and deploying machine-learning methods that detect or locate anomalies in 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. Models are built on Lightning, with inference options through Torch, Lightning, Gradio, or 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, with output options for industrial pipelines using ROS messages or MQTT. Studio is an actively developed pre-release, offered as a Docker container or standalone application, and may be incomplete or unstable. The library is Apache-2.0 licensed and free.
Who it is for
Anomalib suits developers and teams building image or video anomaly detection workflows who want a Python API, CLI, or self-hosted deployment. Studio may suit users seeking a low/no-code web interface, provided they can work with pre-release software.
What is good
- Free, Apache-2.0-licensed library.
- Training, inference, and benchmarking via Python API and CLI.
- Supports image and video anomaly detection.
- Most models export to OpenVINO IR.
- Studio accepts camera feeds or image folders.
What to know first
- Studio is a pre-release and may be unstable.
- Intel GPU training supports only one GPU.
- Intel XPU installation is listed for Linux only.
Verdict
Anomalib combines model development and benchmarking with several inference and deployment routes. Studio adds a visual workflow, but its pre-release status and stated instability are important considerations.
Anomalib plans and pricing
All plansCompared on anomaly detection software
- Detection method
- machine-learning
- Supported data
- images, videos
- Deployment options
- self-hosted
