---
library_name: pytorch
license: other
tags:
- backbone
- bu_auto
- android
pipeline_tag: image-classification

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mnasnet05/web-assets/model_demo.png)

# MNASNet05: Optimized for Qualcomm Devices

MNASNet05 is a machine learning model that can classify images from the Imagenet dataset. It can also be used as a backbone in building more complex models for specific use cases.

This is based on the implementation of MNASNet05 found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/mnasnet.py).
This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/mnasnet05) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).

Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.

## Getting Started
There are two ways to deploy this model on your device:

### Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

| Runtime | Precision | Chipset | SDK Versions | Download |
|---|---|---|---|---|
| ONNX | float | Universal | QAIRT 2.37, ONNX Runtime 1.23.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mnasnet05/releases/v0.46.0/mnasnet05-onnx-float.zip)
| ONNX | w8a16 | Universal | QAIRT 2.37, ONNX Runtime 1.23.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mnasnet05/releases/v0.46.0/mnasnet05-onnx-w8a16.zip)
| QNN_DLC | float | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mnasnet05/releases/v0.46.0/mnasnet05-qnn_dlc-float.zip)
| QNN_DLC | w8a16 | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mnasnet05/releases/v0.46.0/mnasnet05-qnn_dlc-w8a16.zip)
| TFLITE | float | Universal | QAIRT 2.42, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/mnasnet05/releases/v0.46.0/mnasnet05-tflite-float.zip)

For more device-specific assets and performance metrics, visit **[MNASNet05 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/mnasnet05)**.


### Option 2: Export with Custom Configurations

Use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/mnasnet05) Python library to compile and export the model with your own:
- Custom weights (e.g., fine-tuned checkpoints)
- Custom input shapes
- Target device and runtime configurations

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for [MNASNet05 on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/mnasnet05) for usage instructions.

## Model Details

**Model Type:** Model_use_case.image_classification

**Model Stats:**
- Model checkpoint: Imagenet
- Input resolution: 224x224
- Number of parameters: 2.21M
- Model size (float): 8.45 MB
- Model size (w8a16): 2.79 MB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| MNASNet05 | ONNX | float | Snapdragon® X Elite | 0.617 ms | 5 - 5 MB | NPU
| MNASNet05 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.493 ms | 0 - 116 MB | NPU
| MNASNet05 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 0.691 ms | 0 - 12 MB | NPU
| MNASNet05 | ONNX | float | Qualcomm® QCS9075 | 0.97 ms | 1 - 3 MB | NPU
| MNASNet05 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.381 ms | 0 - 99 MB | NPU
| MNASNet05 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.317 ms | 1 - 99 MB | NPU
| MNASNet05 | ONNX | w8a16 | Snapdragon® X Elite | 0.645 ms | 2 - 2 MB | NPU
| MNASNet05 | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.521 ms | 0 - 111 MB | NPU
| MNASNet05 | ONNX | w8a16 | Qualcomm® QCS6490 | 29.306 ms | 8 - 11 MB | CPU
| MNASNet05 | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 0.705 ms | 0 - 44 MB | NPU
| MNASNet05 | ONNX | w8a16 | Qualcomm® QCS9075 | 0.899 ms | 0 - 3 MB | NPU
| MNASNet05 | ONNX | w8a16 | Qualcomm® QCM6690 | 18.804 ms | 9 - 16 MB | CPU
| MNASNet05 | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.368 ms | 0 - 100 MB | NPU
| MNASNet05 | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 13.001 ms | 9 - 16 MB | CPU
| MNASNet05 | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.329 ms | 0 - 100 MB | NPU
| MNASNet05 | QNN_DLC | float | Snapdragon® X Elite | 0.916 ms | 1 - 1 MB | NPU
| MNASNet05 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.516 ms | 0 - 46 MB | NPU
| MNASNet05 | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 2.318 ms | 1 - 30 MB | NPU
| MNASNet05 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 0.775 ms | 1 - 2 MB | NPU
| MNASNet05 | QNN_DLC | float | Qualcomm® SA8775P | 1.091 ms | 1 - 31 MB | NPU
| MNASNet05 | QNN_DLC | float | Qualcomm® QCS9075 | 0.974 ms | 1 - 3 MB | NPU
| MNASNet05 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 1.579 ms | 0 - 48 MB | NPU
| MNASNet05 | QNN_DLC | float | Qualcomm® SA7255P | 2.318 ms | 1 - 30 MB | NPU
| MNASNet05 | QNN_DLC | float | Qualcomm® SA8295P | 1.416 ms | 0 - 28 MB | NPU
| MNASNet05 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.379 ms | 0 - 33 MB | NPU
| MNASNet05 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.291 ms | 1 - 33 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Snapdragon® X Elite | 0.924 ms | 0 - 0 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.53 ms | 0 - 37 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 2.225 ms | 2 - 4 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 1.665 ms | 0 - 26 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 0.785 ms | 0 - 2 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Qualcomm® SA8775P | 4.096 ms | 0 - 27 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 0.925 ms | 0 - 2 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 3.052 ms | 0 - 138 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Qualcomm® QCS8450 (Proxy) | 0.958 ms | 0 - 39 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Qualcomm® SA7255P | 1.665 ms | 0 - 26 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Qualcomm® SA8295P | 1.235 ms | 0 - 23 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.361 ms | 0 - 25 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 0.785 ms | 0 - 24 MB | NPU
| MNASNet05 | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.29 ms | 0 - 29 MB | NPU
| MNASNet05 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.519 ms | 0 - 47 MB | NPU
| MNASNet05 | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 2.331 ms | 0 - 30 MB | NPU
| MNASNet05 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 0.776 ms | 0 - 1 MB | NPU
| MNASNet05 | TFLITE | float | Qualcomm® SA8775P | 1.107 ms | 0 - 33 MB | NPU
| MNASNet05 | TFLITE | float | Qualcomm® QCS9075 | 0.978 ms | 0 - 8 MB | NPU
| MNASNet05 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 1.581 ms | 0 - 49 MB | NPU
| MNASNet05 | TFLITE | float | Qualcomm® SA7255P | 2.331 ms | 0 - 30 MB | NPU
| MNASNet05 | TFLITE | float | Qualcomm® SA8295P | 1.432 ms | 0 - 29 MB | NPU
| MNASNet05 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.384 ms | 0 - 35 MB | NPU
| MNASNet05 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.29 ms | 0 - 34 MB | NPU

## License
* The license for the original implementation of MNASNet05 can be found
  [here](https://github.com/pytorch/vision/blob/main/LICENSE).

## References
* [MnasNet: Platform-Aware Neural Architecture Search for Mobile](https://arxiv.org/abs/1807.11626)
* [Source Model Implementation](https://github.com/pytorch/vision/blob/main/torchvision/models/mnasnet.py)

## Community
* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
* For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).