Netron

Netron is a graphical viewer for neural‑network, deep‑learning and machine‑learning models. It can open model files in a wide range of formats, including ONNX, TensorFlow Lite, PyTorch, Core ML, Keras, Caffe, Darknet, TensorFlow.js, Safetensors and NumPy, with experimental support for TorchScript, MLIR, TensorFlow, OpenVINO, RKNN, ncnn, MNN, PaddlePaddle, GGUF and scikit‑learn. The tool renders the model architecture as an interactive graph, showing layers, operations and connections.

The application is available as a web‑based viewer, native desktop binaries for macOS, Linux and Windows, and a Python package that can be launched from the command line or imported into code. Installation options include direct downloads, package managers such as Homebrew and winget, and pip for Python environments. Users can load a model file locally or via a URL to explore its structure without needing the original training framework.

Netron is open‑source under the MIT license, self‑hostable, and free to use without subscription. It targets developers, researchers and engineers who need to inspect, debug or document model topologies across many machine‑learning ecosystems.