Run AI on-device with Google AI Edge

Our leading end-to-end stack for building and deploying on-device ML and AI across platforms

Jump to next section

Start now with task APIs or deploy custom models

Integrate task APIs into your app

Add prebuilt AI features like background blur, face mesh, and object detection.

MediaPipe

Run LLMs on-device

Run the same LLM across Android, iOS, Web, and embedded devices.

LiteRT-LM

Deploy a custom model across platforms

Convert and deploy hardware-accelerated models from PyTorch, JAX, TensorFlow, and Keras.

LiteRT

TASK APIS

DEPLOY CUSTOM MODELS

The on-device technology that powers Google's apps

Live translation on Pixel

Dynamic backgrounds on Google Meet

Creator effects on YouTube Shorts

Explore the Google AI Edge ecosystem

Your app

MediaPipe

Prebuilt APIs for common AI tasks.

Get started

LiteRT-LM

SDKs to run LLMs across platforms.

Get started

LiteRT

Hardware-accelerated runtimes for custom models.

Get started

CPU

GPU

NPU

Google AI Edge Portal

Benchmark at scale on real Android devices.

Sign up now

Model Explorer

Visualize and debug model architectures.

Explore now

Google AI Edge Quantitizer

Quantize converted LiteRT models.

Start now

Try it now in Google AI Edge Gallery

An interactive app to explore GenAI use cases and try out on-device models.

Recent videos and blogs

On-device GenAI in Chrome, Chromebook Plus, and Pixel Watch with LiteRT-LM

SEPT. 24, 2025 / Mobile

Google AI Edge Gallery: Now with audio and on Google Play

SEPT. 9, 2025 / Mobile

LiteRT: Maximum performance, simplified

MAY 20, 2025 / AI Edge

On-device small language models with multimodality, RAG, and Function Calling

MAY 20, 2025 / AI Edge