netron

Netron is a viewer for neural network, deep learning and machine learning models. It supports a wide range of model formats including ONNX, TensorFlow Lite, PyTorch, torch.export, ExecuTorch, TorchScript, TensorFlow, Core ML, OpenVINO, Keras, Caffe, Darknet, Safetensors, and NumPy. Experimental support is available for MLIR, JAX, GGUF, RKNN, ncnn, MNN, PaddlePaddle, and scikit-learn. ## Installation - **Browser**: Start the browser version at [netron.app](https://netron.app). - **macOS**: Download the `.dmg` file from the [releases page](https://github.com/lutzroeder/netron/releases/latest) or run `brew install --cask netron`. - **Linux**: Download the `.deb` or `.rpm` file from the releases page. - **Windows**: Download the `.exe` installer or run `winget install -s winget netron`. - **Python**: Install via `pip install netron`, then run `netron [FILE]` or `netron.start('[FILE]')`. ## Sample Models The README provides links to sample model files for various formats, which can be downloaded or opened directly in the browser version. Examples include ONNX (squeezenet), TorchScript (traced_online_pred_layer), TensorFlow Lite (yamnet), TensorFlow (chessbot), Keras (mobilenet), MLIR (edge_detection), Core ML (exermote), and Darknet (yolo).