Total runs:
30.4K
Run Growth:
30.4K
Growth Rate:
100.00%
Sponsored by
iMean AI Trip Planner
NexaAI / OmniNeural-4B
Model's Last Updated: November 07 2025
any-to-any
Introduction of OmniNeural-4B
Model Details of OmniNeural-4B
OmniNeural — World’s First NPU-aware Multimodal Model
Overview
OmniNeural is the first fully multimodal model designed specifically for Neural Processing Units (NPUs). It natively understands text, images, and audio , and runs across PCs, mobile devices, automobile, IoT, and robotics.
Demos
📱 Mobile Phone NPU - Demo on Samsung S25 Ultra
The first-ever fully local, multimodal, and conversational AI assistant that hears you and sees what you see, running natively on Snapdragon NPU for long battery life and low latency.
✨ PC NPU - Capabilities Highlights
|
🖼️ Multi-Image Reasoning
|
🎤 Image + Audio → Function Call
|
🎶 Multi-Audio Comparison
|
Key Features
- Multimodal Intelligence – Processes text, image, and audio in a unified model for richer reasoning and perception.
- NPU-Optimized Architecture – Uses ReLU ops, sparse tensors, convolutional layers, and static graph execution for maximum throughput — 20% faster than non-NPU-aware models .
- Hardware-Aware Attention – Attention patterns tuned for NPU, lowering compute and memory demand .
- Native Static Graph – Supports variable-length multimodal inputs with stable, predictable latency .
- Performance Gains – 9× faster audio processing and 3.5× faster image processing on NPUs compared to baseline encoders .
- Privacy-First Inference – All computation stays local: private, offline-capable, and cost-efficient.
Performance / Benchmarks
Human Evaluation (vs baselines)
- Vision : Wins/ties in ~75% of prompts against Apple Foundation, Gemma-3n-E4B, Qwen2.5-Omni-3B.
- Audio : Clear lead over baselines, much better than Gemma3n and Apple foundation model.
- Text : Matches or outperforms leading multimodal baselines.
Nexa Attention Speedups
- 9× faster audio encoding (vs Whisper encoder).
- 3.5× faster image encoding (vs SigLIP encoder).
Architecture Overview
OmniNeural’s design is tightly coupled with NPU hardware:
- NPU-friendly ops (ReLU > GELU/SILU).
- Sparse + small tensor multiplications for efficiency.
- Convolutional layers favored over linear for better NPU parallelization.
- Hardware-aware attention patterns to cut compute cost.
- Static graph execution for predictable latency.
Production Use Cases
-
PC & Mobile – On-device AI agents combine voice, vision, and text for natural, accurate responses.
- Examples: Summarize slides into an email (PC)*, *extract action items from chat (mobile).
- Benefits: Private, offline, battery-efficient.
-
Automotive – In-car assistants handle voice control, cabin safety, and environment awareness .
- Examples: Detects risks (child unbuckled, pet left, loose objects) and road conditions (fog, construction).
- Benefits: Decisions run locally in milliseconds.
-
IoT & Robotics – Multimodal sensing for factories, AR/VR, drones, and robots .
- Examples: Defect detection, technician overlays, hazard spotting mid-flight, natural robot interaction.
- Benefits: Works without network connectivity.
How to use
⚠️ Hardware requirement: OmniNeural-4B currently runs only on Qualcomm NPUs (e.g., Snapdragon-powered AIPC).
Apple NPU support is planned next.
1) Install Nexa-SDK
Download and follow the steps in Nexa's model page:
- Windows on Arm (AIPC) : Download Nexa-SDK
- (Other platforms coming soon)
2) Get an access token
Create a token in the Model Hub, then log in:
nexa config set license '<access_token>'
3) Run the model
Running:
nexa infer NexaAI/OmniNeural-4B
/mic mode. Once the model is running, you can type below to record your voice directly in terminal
> /mic
For images and audio, simply drag your files into the command line. Remember to leave space between file paths.
Links & Community
- Issues / Feedback: Use the HF Discussions tab or submit an issue in our discord or nexa-sdk github.
- Roadmap & updates: Follow us on X and Discord.
If you want to see more NPU-first, multimodal releases on HF, please give our model a like ❤️.
Citation
@misc{
title={OmniNeural: World’s First NPU-aware Multimodal Model},
author={Nexa AI},
year={2025},
url={https://huggingface.co/NexaAI/OmniNeural-4B},
}
Runs of NexaAI OmniNeural-4B on huggingface.co
155
Total runs
0
24-hour runs
10
3-day runs
17
7-day runs
17
30-day runs
More Information About OmniNeural-4B huggingface.co Model
Url of OmniNeural-4B
OmniNeural-4B huggingface.co Url
Provider of OmniNeural-4B huggingface.co
Other API from NexaAI
Updated:October 28 2025
Total runs:
4.6K
Run Growth:
1.5K
Growth Rate:
33.83%
Total runs:
3.2K
Run Growth:
1.5K
Growth Rate:
35.76%
Total runs:
2.4K
Run Growth:
-4.4K
Growth Rate:
-178.72%
Total runs:
1.4K
Run Growth:
37
Growth Rate:
2.66%
Total runs:
1.0K
Run Growth:
-71
Growth Rate:
-6.81%
Total runs:
905
Run Growth:
142
Growth Rate:
15.69%
Total runs:
875
Run Growth:
-221
Growth Rate:
-25.26%
Total runs:
758
Run Growth:
-660
Growth Rate:
-87.07%
Total runs:
523
Run Growth:
-559
Growth Rate:
-106.88%
Total runs:
502
Run Growth:
4
Growth Rate:
0.80%
Total runs:
480
Run Growth:
215
Growth Rate:
44.79%
Total runs:
353
Run Growth:
-895
Growth Rate:
-253.54%
Total runs:
301
Run Growth:
31
Growth Rate:
10.30%
Total runs:
292
Run Growth:
-13
Growth Rate:
-4.45%
Total runs:
148
Run Growth:
63
Growth Rate:
42.57%
Total runs:
145
Run Growth:
6
Growth Rate:
4.14%
Total runs:
126
Run Growth:
34
Growth Rate:
26.98%
Total runs:
92
Run Growth:
-44
Growth Rate:
-47.83%
Total runs:
79
Run Growth:
-38
Growth Rate:
-48.10%
Total runs:
64
Run Growth:
13
Growth Rate:
20.31%
Total runs:
63
Run Growth:
-45
Growth Rate:
-71.43%
Total runs:
57
Run Growth:
-179
Growth Rate:
-314.04%
Total runs:
51
Run Growth:
0
Growth Rate:
0.00%
Total runs:
51
Run Growth:
7
Growth Rate:
13.73%
Total runs:
45
Run Growth:
-336
Growth Rate:
-746.67%
Total runs:
44
Run Growth:
3
Growth Rate:
6.82%
Total runs:
37
Run Growth:
16
Growth Rate:
42.11%
Total runs:
34
Run Growth:
-6
Growth Rate:
-17.65%
Total runs:
34
Run Growth:
-17
Growth Rate:
-50.00%
Total runs:
32
Run Growth:
-40
Growth Rate:
-125.00%
Total runs:
30
Run Growth:
-3
Growth Rate:
-10.00%
Total runs:
29
Run Growth:
0
Growth Rate:
0.00%
Total runs:
28
Run Growth:
-15
Growth Rate:
-53.57%
Total runs:
26
Run Growth:
-2
Growth Rate:
-7.69%
Total runs:
26
Run Growth:
15
Growth Rate:
62.50%
Total runs:
25
Run Growth:
-39
Growth Rate:
-156.00%
Total runs:
24
Run Growth:
-220
Growth Rate:
-916.67%
Total runs:
24
Run Growth:
-6
Growth Rate:
-25.00%
Total runs:
23
Run Growth:
1
Growth Rate:
4.35%
Total runs:
21
Run Growth:
2
Growth Rate:
9.52%
Total runs:
16
Run Growth:
1
Growth Rate:
7.69%
Total runs:
15
Run Growth:
3
Growth Rate:
20.00%
Total runs:
14
Run Growth:
1
Growth Rate:
7.14%
Total runs:
13
Run Growth:
0
Growth Rate:
0.00%
Total runs:
12
Run Growth:
-1
Growth Rate:
-8.33%
Total runs:
12
Run Growth:
4
Growth Rate:
33.33%
Total runs:
11
Run Growth:
1
Growth Rate:
10.00%
Total runs:
11
Run Growth:
-60
Growth Rate:
-545.45%
Total runs:
11
Run Growth:
0
Growth Rate:
0.00%
Total runs:
11
Run Growth:
-12
Growth Rate:
-109.09%
Total runs:
11
Run Growth:
-23
Growth Rate:
-209.09%
Total runs:
11
Run Growth:
1
Growth Rate:
9.09%
Total runs:
10
Run Growth:
4
Growth Rate:
40.00%
Total runs:
10
Run Growth:
7
Growth Rate:
70.00%
Total runs:
9
Run Growth:
-3
Growth Rate:
-33.33%
Total runs:
8
Run Growth:
0
Growth Rate:
0.00%
Total runs:
7
Run Growth:
4
Growth Rate:
57.14%
Total runs:
5
Run Growth:
-11
Growth Rate:
-220.00%
Total runs:
5
Run Growth:
0
Growth Rate:
0.00%
Total runs:
5
Run Growth:
-2
Growth Rate:
-40.00%