converting-pytorch-to-litert
Converts PyTorch models (e.g. ResNet, timm, HuggingFace transformers) directly to LiteRT (.tflite) flatbuffer format. Use when converting PyTorch models to TFLite, setting up export environments, or troubleshooting torch-to-litert conversion bugs. Don't use for ONNX exports or converting existing TensorFlow models.
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Check it deeper on your machine
It has changed since its scan on 17 Nov 2026. Its scripts were only partly scanned.
As a Claude Code plugin, install it and then ask Claude Code:
/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/google-ai-edge/litert-torch with scanskillsafety
Or from a terminal, clone it and run it:
git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/google-ai-edge/litert-torch'
It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.
Badge for your README
Markdown
[](/listings/google-ai-edge--litert-torch)HTML
<a href="/listings/google-ai-edge--litert-torch"><img src="/api/listings/google-ai-edge--litert-torch/badge.svg" alt="findsafeskills safety rating" /></a>Install
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
Requirements and Dependencies
* Python versions: `>=3.10` (**Python 3.11 is highly recommended.**) * Operating system: Linux * PyTorch: [](https://pypi.org/project/torch/) * TensorFlow: [](https://pypi.org/project/tf-nightly/)
Python Virtual Env
Set up a Python virtualenv (we strongly recommend Python 3.11):
python3.11 -m venv --prompt litert-torch venv source venv/bin/activate
The latest stable release can be installed with (we include `torchvision` here to run the quickstart example below, and `ai-edge-litert` for CLI benchmarking tools):
pip install litert-torch torchvision ai-edge-litert
Alternately, the nightly version can be installed with:
pip install --pre litert-torch-nightly torchvision ai-edge-litert litert-cli-nightly
* The list of versioned releases can be seen [here](https://github.com/google-ai-edge/litert-torch/releases). * The full list of PyPi releases (including nightly builds) can be seen [here](https://pypi.org/project/litert-torch/#history).
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
SKILL.md · Manifest · 27 lines
- ---
- name: converting-pytorch-to-litert
- description: >-
- Converts PyTorch models (e.g. ResNet, timm, HuggingFace transformers) directly to
- LiteRT (.tflite) flatbuffer format. Use when converting PyTorch models to TFLite,
- setting up export environments, or troubleshooting torch-to-litert conversion bugs.
- Don't use for ONNX exports or converting existing TensorFlow models.
- ---
- # PyTorch to LiteRT Conversion Guide
- **ATTENTION ALL AGENTS:** If you are tasked with converting a PyTorch model (e.g., ResNet or similar architectures) to a `.tflite` flatbuffer format in this repository, **you must strictly follow these rules based on recent project findings:**
- ## 1. Direct Conversion Only (No ONNX)
- Bypass ONNX entirely. Do not attempt to export the model to ONNX or use `onnx2tf`. Convert directly from PyTorch to LiteRT.
- ## 2. Use `litert-torch`
- Use the `litert-torch` library for the direct PyTorch to TFLite conversion. Note that `ai-edge-torch` is deprecated and has been officially renamed to `litert-torch`.
- ## 3. Strict Python Environment Rules
- - **Do NOT use Python 3.14**. It currently has compatibility issues with typing in `torchao` (`AttributeError: 'typing.Union' object has no attribute '__module__'`).
- - **Use Python 3.11** (e.g., `python3.11`) as it is highly recommended for stability and avoids the aforementioned type hint bugs.
- ## 4. Dependencies
- Before running any conversion script, ensure the following are installed in your `python3.11` environment:
- ```bash
- python3.11 -m pip install torch transformers timm litert-torch ai-edge-litert litert-cli-nightly
README.md · README · 114 lines
- # LiteRT Torch
- LiteRT Torch is a python library that supports converting PyTorch models into a
- .tflite format, which can then be run with [LiteRT](https://ai.google.dev/edge/litert).
- This enables applications for Android, iOS and IOT that can run models
- completely on-device. LiteRT Torch offers broad CPU coverage, with initial GPU
- and NPU support. LiteRT Torch seeks to closely integrate with PyTorch,
- building on top of torch.export() and providing good coverage of Core ATen
- operators.
- To get started converting PyTorch models to LiteRT, see additional details in
- the [PyTorch converter](#pytorch-converter) section. For the particular case of
- Large Language Models (LLMs) and transformer-based models, the [Generative
- API](#generative-api) supports model authoring and quantization to enable
- improved on device performance.
- Although part of the same PyPi package, the PyTorch converter is a Beta release,
- while the Generative API is an Alpha release. Please see the [release
- notes](https://github.com/google-ai-edge/litert-torch/releases/) for additional
- information.
- ## Installation
- ### Requirements and Dependencies
- * Python versions: `>=3.10` (**Python 3.11 is highly recommended.**)
- * Operating system: Linux
- * PyTorch: [](https://pypi.org/project/torch/)
- * TensorFlow: [](https://pypi.org/project/tf-nightly/)
- <!-- requirement badges are updated by ci/update_nightly_versions.py -->
- ### Python Virtual Env
- Set up a Python virtualenv (we strongly recommend Python 3.11):
- ```bash
- python3.11 -m venv --prompt litert-torch venv
- source venv/bin/activate
- ```
- The latest stable release can be installed with (we include `torchvision` here to run the quickstart example below, and `ai-edge-litert` for CLI benchmarking tools):
- ```bash
- pip install litert-torch torchvision ai-edge-litert
- ```
- Alternately, the nightly version can be installed with:
- ```bash
- pip install --pre litert-torch-nightly torchvision ai-edge-litert litert-cli-nightly
- ```
- * The list of versioned releases can be seen [here](https://github.com/google-ai-edge/litert-torch/releases).
- * The full list of PyPi releases (including nightly builds) can be seen [here](https://pypi.org/project/litert-torch/#history).
- ## PyTorch Converter
- Here are the steps needed to convert a PyTorch model to a .tflite flatbuffer:
- ```python
- import torch
- import torchvision
- import litert_torch
- # Use resnet18 with pre-trained weights.
- resnet18 = torchvision.models.resnet18(torchvision.models.ResNet18_Weights.IMAGENET1K_V1)
- with torch.no_grad():
- sample_inputs = (torch.randn(1, 3, 224, 224),)
- # Convert and serialize PyTorch model to a .tflite flatbuffer. Note that we
- # are setting the model to evaluation mode prior to conversion.
- edge_model = litert_torch.convert(resnet18.eval(), sample_inputs)
- edge_model.export("resnet18.tflite")
- ```
- ### Next Steps: Running the Model
- Once exported, you can run this model on-device using the **LiteRT compiled model API** for native C++ or Java deployments, ensuring maximum performance across CPU, GPU, and NPU hardware accelerators.
- The [getting started](docs/pytorch_converter/getting_started.ipynb) Jupyter
- notebook gives an initial walkthrough of the conversion process and can be tried
- out with Google Colab.
- Additional technical details of the PyTorch Converter are [here](docs/pytorch_converter/README.md).
- ## Generative API
- The LiteRT Torch Generative API is a Torch native library for authoring
- mobile-optimized PyTorch Transformer models, which can be converted to LiteRT-LM models,
- allowing users to easily deploy Large Language Models (LLMs) on edge
- devices. Users can run the converted models via [LiteRT-LM](https://github.com/google-ai-edge/LiteRT-LM).
- *Tip: When working with the Generative API, you can package your converted `.tflite` model alongside a tokenizer into a deployment-ready `.litertlm` container using the `litert-lm-builder` CLI tool (installed via the nightly package).*
- More detailed documentation can be found [here](litert_torch/generative).
- The Generative API currently supports CPU and GPU, with planned support for NPU.
- A further future direction is to collaborate with the PyTorch community to
- ensure that frequently used transformer abstractions can be directly supported
- without reauthoring.
- ## Build Status
- Build Type | Status |
- ----------- | --------------|
- Generative API (Linux) | [](https://github.com/google-ai-edge/litert-torch/actions/workflows/nightly_generative_api.yml) |
- Model Coverage (Linux) | [](https://github.com/google-ai-edge/litert-torch/actions/workflows/nightly_model_coverage.yml) |
- Unit tests (Linux) | [](https://github.com/google-ai-edge/litert-torch/actions/workflows/nightly_unittests.yml) |
- Nightly Release | [](https://github.com/google-ai-edge/litert-torch/actions/workflows/nightly_release.yml) |
- # Contributing
- See our [contribution documentation](CONTRIBUTING.md).
- # Getting Help
- Please [create a GitHub issue](https://github.com/google-ai-edge/litert-torch/issues/new/choose) with any questions.
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.