Google AI Edge LiteRT Samples
A collection of official and community sample apps, model recipes and utilities for LiteRT, Google's on-device machine learning framework formerly known as TensorFlow Lite, and LiteRT-LM, its layer for running language models on-device.
This repository demonstrates LiteRT's CompiledModel and legacy Interpreter APIs, its Tensor API, and LiteRT-LM, an orchestration layer for running large language models on-device with maximum performance. Samples cover end-to-end model conversion and deployment pipelines across speech recognition, text-to-speech, multimodal vision-and-language apps, image generation, image segmentation and classification, and NPU acceleration on Qualcomm and Google TPU hardware.
Recent additions include a streaming text-to-speech Android sample, an end-to-end automatic speech recognition sample, a multimodal app combining vision and audio/text generation, Qwen3-TTS and Qwen3 ASR model recipes, and a text-to-image diffusion sample. Agent skills and utilities are also included for model conversion, quantization, on-device verification and app scaffolding, along with a GPU conversion toolkit and shared Kotlin helpers.