Installation

Embedl Deploy is a pure Python package with the only dependency being PyTorch. Install with:

pip install "embedl-deploy[tensorrt]"

Embedl Deploy supports several compilation backends, each installed as an optional extra — pick the one matching your target hardware (e.g. "tensorrt" for NVIDIA GPUs). Omit the extra to install just the base package if you only need to build models in PyTorch and export to ONNX.

For model compilation, you may need additional tools such as TensorRT (trtexec), but these are not required to create deployment-ready models in PyTorch and export them to ONNX.