MLIA LiteRT Converter Plugin

This repository contains the MLIA converter plugin that translates LiteRT / TensorFlow Lite .tflite models into TOSA so they can be consumed by MLIA backends and target flows that operate on TOSA artifacts.

The package is distributed as mlia-converters-litert. When installed, it registers the transformer name tflite_to_tosa with MLIA through the plugin entry-point system.

Table of Contents

• Overview
• Repository contents
• Installation
• How MLIA uses this plugin
• Reporting bugs
• Development (uv)
• Documentation
• Releases
• License
• Trademarks and copyrights

Overview

This plugin provides the conversion bridge between LiteRT / TensorFlow Lite .tflite model files and TOSA-based MLIA backends. It gives the wider MLIA ecosystem a dedicated package for .tflite-to-TOSA conversion and keeps converter logic separate from the core MLIA framework.

The implementation package lives under src/mlia/backend/tosa_converter_for_tflite/ and includes:

• Conversion logic.
• Transformer registration.
• Backend installation metadata used by MLIA.

Repository contents

• src/mlia/backend/tosa_converter_for_tflite/: implementation package and plugin registration.
• tests/: unit tests for converter registration and conversion behaviour.
• pre_commit_hooks/: local repository hooks shared with CI quality checks.
• pyproject.toml: package metadata, including the public tosa-converter-for-tflite dependency.

Installation

Install the package into an environment that already contains mlia:

pip install mlia-converters-litert

For source-based development with uv:

uv sync --group dev

The project supports the Python versions declared in pyproject.toml and keeps its direct dependency surface small, with MLIA as the primary runtime dependency.

How MLIA uses this plugin

MLIA discovers this repository through the mlia.plugin.transformer entry point. When installed, the plugin registers the transformer name tflite_to_tosa.

The package and transformer use different names:

• tflite_to_tosa is the transformer name used in MLIA workflows.
• tosa_converter_for_tflite is the Python implementation package.

That means downstream MLIA components can:

• Discover the converter without hard-coded import paths.
• Request a LiteRT / TensorFlow Lite .tflite-to-TOSA conversion through the transformer registry.
• Treat the converter as a separately versioned plugin package.

For more implementation detail, see docs/README.md.

Reporting bugs

Report bugs by creating GitHub issues. Use the arm/mlia issue tracker by default.

Only open an issue in arm/mlia-converters-litert when the bug is clearly and specifically in this LiteRT converter plugin.

Development (uv)

This repository uses uv for environment management and test execution. Ensure the Python version in .python-version is available, then install dependencies:

uv sync --group dev

Run unit tests (uses dependencies installed from the package index, including mlia):

uv run pytest --no-success-flaky-report tests/

Run a quick test subset:

uv run pytest --no-success-flaky-report -m "not slow" tests/

Lint checks:

uv run pre-commit run --all-files

Build a wheel:

uv build --wheel

Documentation

Additional repository documentation lives in docs/README.md.

Releases

Latest changes and release history can be found in MLIA LiteRT Converter releases.

License

This project is licensed under the Apache License 2.0. See LICENSES/Apache-2.0.txt for the full license text.

Trademarks and copyrights

• Arm is a registered trademark or trademark of Arm Limited (or its subsidiaries) in the U.S. and/or elsewhere.
• TensorFlow is a trademark of Google LLC.
• Linux is the registered trademark of Linus Torvalds in the U.S. and elsewhere.
• Python is a registered trademark of the PSF.