Update README.md
Browse files
README.md
CHANGED
@@ -119,23 +119,59 @@ OmniNeural’s design is tightly coupled with NPU hardware:
|
|
119 |
|
120 |
---
|
121 |
|
122 |
-
##
|
123 |
|
124 |
-
> ⚠️
|
125 |
-
> Apple NPU support is planned
|
126 |
|
127 |
-
|
|
|
128 |
|
129 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
130 |
|
131 |
## **License & Citation**
|
132 |
|
133 |
```bibtex
|
134 |
@misc{
|
135 |
-
title={OmniNeural: NPU-
|
136 |
author={Nexa AI},
|
137 |
year={2025},
|
138 |
-
primaryClass={cs.CV},
|
139 |
url={https://huggingface.co/NexaAI/OmniNeural-4B},
|
140 |
}
|
141 |
```
|
|
|
119 |
|
120 |
---
|
121 |
|
122 |
+
## How to use
|
123 |
|
124 |
+
> ⚠️ **Hardware requirement:** OmniNeural-4B currently runs **only on Qualcomm NPUs** (e.g., Snapdragon-powered AIPC).
|
125 |
+
> Apple NPU support is planned next.
|
126 |
|
127 |
+
### 1) Install Nexa-SDK
|
128 |
+
Download and follow the steps in Nexa's model page:
|
129 |
|
130 |
+
- **Windows on Arm (AIPC)**: [Download Nexa-SDK](https://sdk.nexa.ai/model)
|
131 |
+
- (Other platforms coming soon)
|
132 |
+
|
133 |
+
### 2) Get an access token
|
134 |
+
Create a token in the Model Hub, then log in:
|
135 |
+
|
136 |
+
```bash
|
137 |
+
nexa config set license '<access_token>'
|
138 |
+
```
|
139 |
+
|
140 |
+
### 3) Run the model
|
141 |
+
Running:
|
142 |
+
|
143 |
+
```bash
|
144 |
+
nexa infer NexaAI/OmniNeural-4B
|
145 |
+
```
|
146 |
+
|
147 |
+
/mic mode. Once the model is running, you can type below to record your voice directly in terminal
|
148 |
+
```bash
|
149 |
+
> /mic
|
150 |
+
```
|
151 |
+
|
152 |
+
For images and audio, simply drag your files into the command line. Remember to leave space between file paths.
|
153 |
+
|
154 |
+
|
155 |
+
## Links & Community
|
156 |
+
|
157 |
+
[](https://discord.com/invite/nexa-ai)
|
158 |
+
|
159 |
+
[](https://x.com/nexa_ai)
|
160 |
+
|
161 |
+
[](https://nexa.ai)
|
162 |
+
|
163 |
+
- **Issues / Feedback:** Use the **HF Discussions** tab or submit an issue in our discord or nexa-sdk github.
|
164 |
+
- **Roadmap & updates:** Follow us on X and Discord.
|
165 |
+
|
166 |
+
> If you want to see more **NPU-first, multimodal** releases on HF, please give our model a like.
|
167 |
|
168 |
## **License & Citation**
|
169 |
|
170 |
```bibtex
|
171 |
@misc{
|
172 |
+
title={OmniNeural: World’s First NPU-aware Multimodal Model},
|
173 |
author={Nexa AI},
|
174 |
year={2025},
|
|
|
175 |
url={https://huggingface.co/NexaAI/OmniNeural-4B},
|
176 |
}
|
177 |
```
|