Skip to content

Release v2.0.0

Choose a tag to compare

@github-actions github-actions released this 12 Jun 20:47
2679653

This release provides the standalone Model Compiler package for the SiMa.ai Neat SDK 2.0.0 workflow. It is functionally compatible with the legacy Model SDK 2.0.0 release, so developers who previously used the Model SDK 2.0.0 compiler tools can use this package for the same model compilation workflows while adopting the newer Neat SDK installation flow.

Platform Support

This release supports the x86/amd64 platform only. It is intended for compatible Ubuntu 22.04 host environments and for x86-based Neat SDK 2.0.0 development setups. ARM/Modalix target installation is not supported for this model compiler release.

Installation

Install the Model Compiler package with sima-cli:

sima-cli install -v 2.0.0 tools/model-compiler/amd64

When running sima-cli sdk setup for the Neat SDK 2.0.0 environment, sima-cli will prompt you to install this Model Compiler version. Installing from inside the Neat SDK requires sima-cli 2.1.11 or later.

Activate the Compiler Environment

After installation, start a new shell session or reload your shell configuration so the activation commands are available:

source ~/.bashrc

Then activate the Model Compiler environment before running compiler tools:

activate-model-compiler

To leave the compiler environment, run:

deactivate-model-compiler

Validation

This release was smoke-tested with the following flows:

  • Installed independently from the Neat SDK 2.0.0 environment and ran llima-compile and mla-masm to confirm the tools launch successfully.
  • Installed independently on an Ubuntu 22.04 host environment compatible with this Model Compiler release.

Notes for Legacy Model SDK 2.0.0 Users

The compiler functionality is compatible with the legacy Model SDK 2.0.0 release. The main difference is packaging and environment management: install the compiler through sima-cli, activate it with activate-model-compiler, and deactivate it with deactivate-model-compiler when finished.