Apache TVM is an open source ML compiler framework for CPUs, GPUs, FPGA’s, microcontrollers, and specialized accelerators. It supports PyTorch, Tensorflow, MXNet, Darket, and a few other frameworks. The project goal is to enable engineers to optimize and run computations efficiently on any hardware backend.

Project Background

• • Compiler: Apache TVM
  • Author: OctoML
  • Initial Release: October 2020
  • Type: Compiler framework for CPUs and GPUs
  • License: Apache License 2.0
  • Language: Python, C++, and others
  • GitHub: apache/tvm has 7.3k stars, 8 releases, and 614 contributors
  • Community: tvm
  • Twitter: @apachetvm