> ## 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.

# NPU Models

> Optimized format for running models on Neural Processing Units, enabling efficient, low-latency inference on edge and specialized hardware.

## 🖥️ **Supported Devices**

NPU models are supported on Qualcomm Neural Processing Units (NPUs):

* **Nexa CLI**: Qualcomm Snapdragon NPU PC
* **NexaML**: Any Qualcomm NPU device ([contact us](https://sdk.nexa.ai/) to request access)

## ⚙️ **Prerequisites**

* If you haven't already, install the nexa-SDK.
* All NPU models require an access token before usage:
  * **Create an account or log in** at [sdk.nexa.ai](https://sdk.nexa.ai)
  * **Generate a token**: Go to **Deployment → Create Token**
  * **Activate your SDK**: Run the following command on the terminal to set your license:
  ```bash bash theme={"dark"}
  nexa config set license '<your_token_here>'
  ```

## **LLM - Language Models**

📝 Language models in NPU format. Try out this quick example:

Try it out:

```bash bash theme={"dark"}
nexa infer NexaAI/qwen3-4B-npu
```

<Check>
  ⌨️  Once model loads, type or paste multi-line text directly into the CLI to chat with the model.
</Check>

***

## **LMM - Multimodal Models**

🖼️ Language models that also accept vision and/or audio inputs. LMM in NPU formats. Try out this quick example:

```bash bash theme={"dark"}
nexa infer NexaAI/OmniNeural-4B
```

<Check>
  ⌨️  Drag photos or audio clips directly into the CLI — you can even drop multiple images at once!
</Check>

## **Supported Model List**

We curated a list of top, high quality models in NPU format.

<Card icon="computer" title="Models for Qualcomm NPU" href="https://huggingface.co/collections/NexaAI/qualcomm-npu-68a56499963641acb2f1eab7" cta="View Huggingface Collection 🤗" />

For more advanced models, you may visit the [Nexa Model Hub](https://sdk.nexa.ai/model). Also, access token is required to download and use these models. To get access token, You can check the **Prerequisites** section of this page.

## 🙋 **Request New Models**

Missing a model? Vote for it on the **[Nexa Wishlist](/en/community/wishlist)** — we build the most-voted models fast!

You can also submit an issue on the <a href="https://github.com/NexaAI/nexa-sdk/issues" target="_blank">nexa-sdk GitHub</a> or request in our Discord/Slack community.

<br />

<Feedback />