Developer Tools

Intel provides a range of developer tools which cover a wide range of topics from heterogeneous computing to deep learning optimization, to help you build and optimize Humanoid Toolkit models and workflows.

For Humanoid workflows, complete the platform getting-started guide before installing these tools.

Please get more information in the following sections:

• Intel® OpenVINO™ is an open-source toolkit for optimizing and deploying deep learning models.

• Intel® Extension for PyTorch* is a library that extends PyTorch* with the latest performance optimizations for Intel hardware.

• Intel® LLM Library for PyTorch* is an LLM optimization library which accelerates local LLM inference and fine-tuning on Intel hardware.

• Intel® oneAPI is a unified programming model that enables developers to write code that can be executed on a variety of hardware accelerators.

• Intel® Extension for OpenXLA* is an extension for OpenXLA* that provides seamless run of JAX models on Intel GPU.

Also, you can find some advice for arranging heterogeneous computing through different workloads in the following section:

• Heterogeneous Computing

Some of the models used in Humanoid Toolkit solutions are enabled on Intel platforms; see the following tutorials:

• OpenVINO Model Guidance