> ## Documentation Index
> Fetch the complete documentation index at: https://docs.nexa.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Overview

> Start here to set up and explore the Nexa SDK for running latest models on Android devices.

## **Android SDK (Beta)**

<img className="block dark:hidden" src="https://mintcdn.com/nexaai/0Bh2MiZtOacPTTJ2/Mintlify-image/Android.jpeg?fit=max&auto=format&n=0Bh2MiZtOacPTTJ2&q=85&s=952427830e04006fcf50c7f293d075b6" alt="Nexa Android SDK" width="1920" height="1080" data-path="Mintlify-image/Android.jpeg" />

<img className="hidden dark:block" src="https://mintcdn.com/nexaai/0Bh2MiZtOacPTTJ2/Mintlify-image/Android.jpeg?fit=max&auto=format&n=0Bh2MiZtOacPTTJ2&q=85&s=952427830e04006fcf50c7f293d075b6" alt="Nexa Android SDK" width="1920" height="1080" data-path="Mintlify-image/Android.jpeg" />

NexaSDK for Android enables on-device AI inference for Android applications. It built upon Nexa AI's proprietary NexaML engine and enables developers to run Large Language Models (LLMs), Vision-Language Models (VLMs), Embeddings, Speech Recognition (ASR), Reranking, and Computer Vision models directly on Android devices with support for NPU, GPU, and CPU inference.

## **Why NPU on mobile**

<iframe className="w-full aspect-video rounded-xl" src="https://www.youtube.com/embed/Qvdtqq-0mqA" title="Why NPU on Mobile" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowFullScreen />

## **Key Features**

* **Multiple Model Types**: Support for LLM, VLM, Embeddings, ASR, Reranker, and Computer Vision models
* **Hardware Acceleration**: CPU, GPU, and NPU (Qualcomm Hexagon NPU) support
* **Easy Integration**: Simple Kotlin/Java API with builder pattern

## **Supported Models**

The following models are supported on Android ARM64 with NPU acceleration :

| Modality      | Model                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    |
| ------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **LLM**       | [`NexaAI/Qwen3-4B-Instruct-2507`](https://huggingface.co/NexaAI/Qwen3-4B-Instruct-2507-npu-mobile)<br />[`NexaAI/Llama3.2-3B-NPU-Turbo`](https://huggingface.co/NexaAI/Llama3.2-3B-NPU-Turbo-NPU-mobile)<br />[`NexaAI/Granite-4-Micro`](https://huggingface.co/NexaAI/Granite-4-Micro-NPU-mobile)<br />[`NexaAI/phi3.5-mini`](https://huggingface.co/NexaAI/phi3.5-mini-npu-mobile)<br />[`NexaAI/phi4-mini`](https://huggingface.co/NexaAI/phi4-mini-npu-mobile)<br />[`NexaAI/LFM2-1.2B-npu`](https://huggingface.co/NexaAI/LFM2-1.2B-npu-mobile)     |
| **VLM**       | [`NexaAI/OmniNeural-4B`](https://huggingface.co/NexaAI/OmniNeural-4B-mobile)                                                                                                                                                                                                                                                                                                                                                                                                                                                                             |
| **Embedding** | [`NexaAI/embeddinggemma-300m-npu`](https://huggingface.co/NexaAI/embeddinggemma-300m-npu-mobile)                                                                                                                                                                                                                                                                                                                                                                                                                                                         |
| **ASR**       | [`NexaAI/parakeet-tdt-0.6b-v3-npu`](https://huggingface.co/NexaAI/parakeet-tdt-0.6b-v3-npu-mobile)                                                                                                                                                                                                                                                                                                                                                                                                                                                       |
| **CV**        | [`NexaAI/paddleocr`](https://huggingface.co/NexaAI/paddleocr-npu-mobile)<br />[`NexaAI/yolo26x`](https://huggingface.co/NexaAI/yolo26x-npu-mobile)<br />[`NexaAI/yolo26l`](https://huggingface.co/NexaAI/yolo26l-npu-mobile)<br />[`NexaAI/yolo26m`](https://huggingface.co/NexaAI/yolo26m-npu-mobile)<br />[`NexaAI/yolo26s`](https://huggingface.co/NexaAI/yolo26s-npu-mobile)<br />[`NexaAI/yolo26n`](https://huggingface.co/NexaAI/yolo26n-npu-mobile)<br />[`NexaAI/depth-anything-v2`](https://huggingface.co/NexaAI/depth-anything-v2-npu-mobile) |

<Note>
  Beyond the NEXA-optimized models listed above, **any GGUF model from the community** can also run on Qualcomm Hexagon NPU. Use the `cpu_gpu` plugin and set `device_id = "dev0"`, powered by the GGML Hexagon backend.
</Note>

## **Get Started**

<Card title="Quickstart for Android" href="/en/nexa-sdk-android/quickstart" icon="rocket">
  Start building with Nexa Android SDK in minutes. Follow our step-by-step guide to integrate AI models into your Android application.
</Card>

<br />

<div class="feedback-wrapper">
  <span class="feedback-label">Was this page helpful?</span>

  <div class="feedback-toggle">
    <input type="radio" name="feedback" id="feedback-yes" class="feedback-input" />

    <label for="feedback-yes" class="feedback-button">
      <img src="https://mintcdn.com/nexaai/g8-zBYnunEyVtcK3/Images/FeedBack/thumbs-up.svg?fit=max&auto=format&n=g8-zBYnunEyVtcK3&q=85&s=0b57c51c8db9940403e7552956e5c30e" alt="Thumbs up" class="feedback-icon" noZoom width="14" height="14" data-path="Images/FeedBack/thumbs-up.svg" />

      Yes
    </label>

    <input type="radio" name="feedback" id="feedback-no" class="feedback-input" />

    <label for="feedback-no" class="feedback-button">
      <img src="https://mintcdn.com/nexaai/g8-zBYnunEyVtcK3/Images/FeedBack/thumbs-down.svg?fit=max&auto=format&n=g8-zBYnunEyVtcK3&q=85&s=ebacf61d57c8259c6df243d329b548b3" alt="Thumbs down" class="feedback-icon" noZoom width="14" height="14" data-path="Images/FeedBack/thumbs-down.svg" />

      No
    </label>
  </div>
</div>