Convert PyTorch GenAI models for on-device inference
The AI Edge Torch Generative API is a
high-performance library designed for authoring and converting
transformer-based PyTorch models into the LiteRT/LiteRT-LM format. This
enables developers to seamlessly deploy generative AI models, specifically
Large Language Models (LLMs), for on-device text and image generation
with ease.
The Torch Generative API supports model conversion for CPU and GPU execution,
with NPU support in development. By pairing Torch Generative API with
LiteRT-LM, you can build responsive, privacy-focused applications
that run generative models entirely on-device.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-12-22 UTC."],[],[]]
