google-ai-edge/litert-torch
Google's tool for converting PyTorch models to LiteRT's .tflite format for on-device ML inference.
Not currently ranked — collecting fresh signals.
star history
LiteRT Torch is a Python library that converts PyTorch models into TFLite flatbuffers for edge deployment. It integrates with torch.export() and provides broad CPU coverage with initial GPU and NPU support for on-device inference. The Generative API specifically supports authoring and quantizing LLMs and transformer-based models to optimize performance on mobile and IoT devices.
Frequently asked
- What is google-ai-edge/litert-torch?
- Google's tool for converting PyTorch models to LiteRT's .tflite format for on-device ML inference.
- Is litert-torch open source?
- Yes — google-ai-edge/litert-torch is open source, released under the Apache-2.0 license.
- What language is litert-torch written in?
- google-ai-edge/litert-torch is primarily written in Jupyter Notebook.
- How popular is litert-torch?
- google-ai-edge/litert-torch has 1.1k stars on GitHub.
- Where can I find litert-torch?
- google-ai-edge/litert-torch is on GitHub at https://github.com/google-ai-edge/litert-torch.