---
id: tflite-runtime
version: "2.14.0"
license: Apache 2.0
license_treatment: permissive
maintenance: dormant
---

# tflite-runtime — TensorFlow Lite is for mobile and embedded devices.

License: permissive · Maintenance: dormant · Downloads: 120.7K/mo

## What it is and what it does

TensorFlow Lite runtime is the official inference engine for running pre-trained machine learning models on mobile and embedded devices. It is designed to execute models with minimal latency and a small footprint, supporting Android, iOS, and other operating systems. The package provides Python bindings to load and run .tflite model files, making it suitable for edge deployment scenarios where full TensorFlow would be too heavy.

The package depends only on numpy and ships as platform-specific wheels for common architectures (x86_64, aarch64, armv7l) across Python 3.8–3.11. It is classified as Production/Stable and carries no known security vulnerabilities, though the package has been dormant since its last release on 2023-10-03.

Use it for:

- Run pre-trained image classification or object detection models on Raspberry Pi or ARM-based edge devices.
- Deploy real-time inference pipelines on mobile devices without the overhead of the full TensorFlow library.
- Execute quantized or pruned models for resource-constrained IoT and embedded systems.
- Integrate lightweight neural networks into Python applications on ARM architectures (aarch64, armv7l).
- Prototype and test model inference performance on edge hardware before production deployment.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

TensorFlow Lite runtime enables on-device machine learning inference on mobile and embedded devices with low latency and minimal binary size.

Yes, if you have a pre-converted .tflite model and need to run inference on mobile or embedded devices. The package is stable, lightweight, and carries no known vulnerabilities. However, the dormant maintenance status means no recent updates or security patches; verify compatibility with your target model format and hardware before committing to production use.

## Install

pip install tflite-runtime
uv add tflite-runtime
poetry add tflite-runtime

## Installing tflite-runtime

Before you install: Medium install friction due to platform-specific wheels for x86_64, aarch64, and armv7l architectures across Python 3.8–3.11. Package is dormant (last release 1046 days ago), but carries Production/Stable status and no known vulnerabilities.

License in practice: Apache 2.0 permissive license allows commercial and private use with minimal restrictions; suitable for most deployment scenarios.

Quickstart:

pip install tflite-runtime
import tflite_runtime.interpreter as tflite
interpreter = tflite.Interpreter(model_path='model.tflite')
interpreter.allocate_tensors()
output = interpreter.get_tensor(interpreter.get_output_details()[0]['index'])

Requires a pre-converted TensorFlow Lite model (.tflite file); numpy is a runtime dependency.

Verify before relying:

- Whether dormant status affects compatibility with current model formats or ecosystem changes.
- Performance characteristics compared to other edge inference runtimes on specific hardware.
- Support timeline and security patch policy for the dormant package.
- Whether the package works on Python versions beyond 3.11.

## Package facts

- License: Apache 2.0 (permissive)
- Python support: unspecified
- Install friction: medium
- Maintenance: dormant
- Downloads: 120.7K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

tensorflow lite inference, on-device machine learning, mobile ml runtime, embedded neural network, tflite model execution, low-latency inference, edge device ml, edge-inference, model-deployment, embedded-ml

[View on SkillFed](https://skillfed.io/packages/tflite-runtime) · [View on PyPI](https://pypi.org/project/tflite-runtime/)