Model Packaging & Acceleration

  • We have upgraded our Package command so that you can see benchmarks that are automatically initiated with every packaging request for different runtimes like TVM, ONNX, and other native runtimes (e.g. TensorFlow/TFLite/PyTorch).

  • Note: Each one of these packaging workflows, including Native Runtime, will produce docker images that allow you to get to deployment quickly.

Did this answer your question?