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

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/inception_v3/web-assets/model_demo.png)

# Inception-v3: Optimized for Qualcomm Devices

InceptionNetV3 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 Inception-v3 found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/inception.py).
This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/inception_v3) 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 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/inception_v3/releases/v0.48.0/inception_v3-onnx-float.zip)
| ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/inception_v3/releases/v0.48.0/inception_v3-onnx-w8a8.zip)
| QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/inception_v3/releases/v0.48.0/inception_v3-qnn_dlc-float.zip)
| QNN_DLC | w8a8 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/inception_v3/releases/v0.48.0/inception_v3-qnn_dlc-w8a8.zip)
| TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/inception_v3/releases/v0.48.0/inception_v3-tflite-float.zip)
| TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/inception_v3/releases/v0.48.0/inception_v3-tflite-w8a8.zip)

For more device-specific assets and performance metrics, visit **[Inception-v3 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/inception_v3)**.


### Option 2: Export with Custom Configurations

Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/inception_v3) 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 [Inception-v3 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/inception_v3) for usage instructions.

## Model Details

**Model Type:** Model_use_case.image_classification

**Model Stats:**
- Model checkpoint: Imagenet
- Input resolution: 224x224
- Number of parameters: 23.9M
- Model size (float): 90.9 MB
- Model size (w8a8): 23.3 MB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| Inception-v3 | ONNX | float | Snapdragon® X2 Elite | 0.691 ms | 46 - 46 MB | NPU
| Inception-v3 | ONNX | float | Snapdragon® X Elite | 1.518 ms | 46 - 46 MB | NPU
| Inception-v3 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1.051 ms | 0 - 60 MB | NPU
| Inception-v3 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 1.483 ms | 0 - 49 MB | NPU
| Inception-v3 | ONNX | float | Qualcomm® QCS9075 | 2.177 ms | 0 - 4 MB | NPU
| Inception-v3 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.925 ms | 0 - 34 MB | NPU
| Inception-v3 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.805 ms | 0 - 39 MB | NPU
| Inception-v3 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.291 ms | 23 - 23 MB | NPU
| Inception-v3 | ONNX | w8a8 | Snapdragon® X Elite | 0.729 ms | 23 - 23 MB | NPU
| Inception-v3 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.506 ms | 0 - 95 MB | NPU
| Inception-v3 | ONNX | w8a8 | Qualcomm® QCS6490 | 25.099 ms | 10 - 39 MB | CPU
| Inception-v3 | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.66 ms | 0 - 26 MB | NPU
| Inception-v3 | ONNX | w8a8 | Qualcomm® QCS9075 | 0.746 ms | 0 - 3 MB | NPU
| Inception-v3 | ONNX | w8a8 | Qualcomm® QCM6690 | 17.931 ms | 10 - 19 MB | CPU
| Inception-v3 | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.431 ms | 0 - 58 MB | NPU
| Inception-v3 | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 14.066 ms | 10 - 21 MB | CPU
| Inception-v3 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.412 ms | 0 - 57 MB | NPU
| Inception-v3 | QNN_DLC | float | Snapdragon® X2 Elite | 0.78 ms | 1 - 1 MB | NPU
| Inception-v3 | QNN_DLC | float | Snapdragon® X Elite | 1.417 ms | 1 - 1 MB | NPU
| Inception-v3 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.02 ms | 0 - 57 MB | NPU
| Inception-v3 | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 7.828 ms | 1 - 33 MB | NPU
| Inception-v3 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 1.368 ms | 1 - 164 MB | NPU
| Inception-v3 | QNN_DLC | float | Qualcomm® SA8775P | 2.247 ms | 1 - 35 MB | NPU
| Inception-v3 | QNN_DLC | float | Qualcomm® QCS9075 | 2.056 ms | 1 - 3 MB | NPU
| Inception-v3 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 2.348 ms | 0 - 55 MB | NPU
| Inception-v3 | QNN_DLC | float | Qualcomm® SA7255P | 7.828 ms | 1 - 33 MB | NPU
| Inception-v3 | QNN_DLC | float | Qualcomm® SA8295P | 2.688 ms | 1 - 31 MB | NPU
| Inception-v3 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.879 ms | 1 - 35 MB | NPU
| Inception-v3 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.758 ms | 1 - 37 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.343 ms | 0 - 0 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.661 ms | 0 - 0 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.458 ms | 0 - 73 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 2.289 ms | 2 - 4 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 1.548 ms | 0 - 47 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.613 ms | 0 - 111 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.78 ms | 0 - 51 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 0.625 ms | 2 - 4 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 5.697 ms | 0 - 50 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.902 ms | 0 - 74 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 1.548 ms | 0 - 47 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 1.157 ms | 0 - 46 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.425 ms | 0 - 51 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.978 ms | 0 - 49 MB | NPU
| Inception-v3 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.37 ms | 0 - 51 MB | NPU
| Inception-v3 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.018 ms | 0 - 127 MB | NPU
| Inception-v3 | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 7.849 ms | 0 - 73 MB | NPU
| Inception-v3 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 1.367 ms | 0 - 132 MB | NPU
| Inception-v3 | TFLITE | float | Qualcomm® SA8775P | 2.265 ms | 0 - 74 MB | NPU
| Inception-v3 | TFLITE | float | Qualcomm® QCS9075 | 2.057 ms | 0 - 48 MB | NPU
| Inception-v3 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 2.35 ms | 0 - 124 MB | NPU
| Inception-v3 | TFLITE | float | Qualcomm® SA7255P | 7.849 ms | 0 - 73 MB | NPU
| Inception-v3 | TFLITE | float | Qualcomm® SA8295P | 2.694 ms | 0 - 70 MB | NPU
| Inception-v3 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.886 ms | 0 - 76 MB | NPU
| Inception-v3 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.758 ms | 0 - 76 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.485 ms | 0 - 75 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Qualcomm® QCS6490 | 2.3 ms | 0 - 25 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 1.677 ms | 0 - 46 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.643 ms | 0 - 1 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Qualcomm® SA8775P | 0.878 ms | 0 - 50 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Qualcomm® QCS9075 | 0.704 ms | 0 - 25 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Qualcomm® QCM6690 | 5.836 ms | 0 - 50 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.951 ms | 0 - 76 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Qualcomm® SA7255P | 1.677 ms | 0 - 46 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Qualcomm® SA8295P | 1.296 ms | 0 - 45 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.416 ms | 0 - 45 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.023 ms | 0 - 49 MB | NPU
| Inception-v3 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.38 ms | 0 - 49 MB | NPU

## License
* The license for the original implementation of Inception-v3 can be found
  [here](https://github.com/pytorch/vision/blob/main/LICENSE).

## References
* [Rethinking the Inception Architecture for Computer Vision](http://arxiv.org/abs/1512.00567)
* [Source Model Implementation](https://github.com/pytorch/vision/blob/main/torchvision/models/inception.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).