---
library_name: pytorch
license: other
tags:
- real_time
- android
pipeline_tag: object-detection

---

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

# MediaPipe-Face-Detection: Optimized for Qualcomm Devices

Designed for sub-millisecond processing, this model predicts bounding boxes and pose skeletons (left eye, right eye, nose tip, mouth, left eye tragion, and right eye tragion) of faces in an image.

This is based on the implementation of MediaPipe-Face-Detection found [here](https://github.com/zmurez/MediaPipePyTorch/).
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/mediapipe_face) 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/mediapipe_face/releases/v0.47.0/mediapipe_face-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/mediapipe_face/releases/v0.47.0/mediapipe_face-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/mediapipe_face/releases/v0.47.0/mediapipe_face-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/mediapipe_face/releases/v0.47.0/mediapipe_face-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/mediapipe_face/releases/v0.47.0/mediapipe_face-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/mediapipe_face/releases/v0.47.0/mediapipe_face-tflite-w8a8.zip)

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


### 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/mediapipe_face) 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 [MediaPipe-Face-Detection on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/mediapipe_face) for usage instructions.

## Model Details

**Model Type:** Model_use_case.object_detection

**Model Stats:**
- Input resolution: 256x256
- Number of output classes: 6
- Number of parameters (FaceDetector): 135K
- Model size (FaceDetector) (w8a8): 260 KB
- Number of parameters (FaceLandmarkDetector): 603K
- Model size (FaceLandmarkDetector) (w8a8): 750 KB
- Model size (FaceDetector) (float): 557 KB
- Model size (FaceLandmarkDetector) (float): 2.33 MB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| FaceDetector | ONNX | float | Snapdragon® X Elite | 0.871 ms | 2 - 2 MB | NPU
| FaceDetector | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.493 ms | 0 - 43 MB | NPU
| FaceDetector | ONNX | float | Qualcomm® QCS8550 (Proxy) | 0.679 ms | 1 - 9 MB | NPU
| FaceDetector | ONNX | float | Qualcomm® QCS9075 | 1.144 ms | 1 - 3 MB | NPU
| FaceDetector | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.413 ms | 0 - 29 MB | NPU
| FaceDetector | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.334 ms | 0 - 34 MB | NPU
| FaceDetector | ONNX | float | Snapdragon® X2 Elite | 0.387 ms | 0 - 0 MB | NPU
| FaceDetector | ONNX | w8a8 | Snapdragon® X Elite | 0.412 ms | 1 - 1 MB | NPU
| FaceDetector | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.193 ms | 0 - 43 MB | NPU
| FaceDetector | ONNX | w8a8 | Qualcomm® QCS6490 | 9.938 ms | 7 - 11 MB | CPU
| FaceDetector | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.282 ms | 0 - 3 MB | NPU
| FaceDetector | ONNX | w8a8 | Qualcomm® QCS9075 | 0.429 ms | 0 - 3 MB | NPU
| FaceDetector | ONNX | w8a8 | Qualcomm® QCM6690 | 7.273 ms | 5 - 13 MB | CPU
| FaceDetector | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.167 ms | 0 - 37 MB | NPU
| FaceDetector | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 5.505 ms | 7 - 16 MB | CPU
| FaceDetector | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.149 ms | 0 - 36 MB | NPU
| FaceDetector | ONNX | w8a8 | Snapdragon® X2 Elite | 0.162 ms | 0 - 0 MB | NPU
| FaceDetector | QNN_DLC | float | Snapdragon® X Elite | 0.691 ms | 1 - 1 MB | NPU
| FaceDetector | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.394 ms | 0 - 42 MB | NPU
| FaceDetector | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 2.908 ms | 1 - 30 MB | NPU
| FaceDetector | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 0.538 ms | 1 - 3 MB | NPU
| FaceDetector | QNN_DLC | float | Qualcomm® SA8775P | 1.021 ms | 0 - 31 MB | NPU
| FaceDetector | QNN_DLC | float | Qualcomm® QCS9075 | 0.891 ms | 1 - 3 MB | NPU
| FaceDetector | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 0.764 ms | 0 - 37 MB | NPU
| FaceDetector | QNN_DLC | float | Qualcomm® SA7255P | 2.908 ms | 1 - 30 MB | NPU
| FaceDetector | QNN_DLC | float | Qualcomm® SA8295P | 1.107 ms | 0 - 23 MB | NPU
| FaceDetector | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.334 ms | 0 - 33 MB | NPU
| FaceDetector | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.276 ms | 1 - 33 MB | NPU
| FaceDetector | QNN_DLC | float | Snapdragon® X2 Elite | 0.482 ms | 1 - 1 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.348 ms | 0 - 0 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.169 ms | 0 - 36 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 0.763 ms | 0 - 2 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 0.684 ms | 0 - 28 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.25 ms | 0 - 16 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Qualcomm® SA8775P | 1.545 ms | 0 - 28 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 0.34 ms | 2 - 4 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 1.278 ms | 0 - 27 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.292 ms | 0 - 38 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Qualcomm® SA7255P | 0.684 ms | 0 - 28 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Qualcomm® SA8295P | 0.594 ms | 0 - 25 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.143 ms | 0 - 27 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.254 ms | 0 - 26 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.136 ms | 0 - 30 MB | NPU
| FaceDetector | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.23 ms | 0 - 0 MB | NPU
| FaceDetector | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.4 ms | 0 - 40 MB | NPU
| FaceDetector | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 2.946 ms | 0 - 28 MB | NPU
| FaceDetector | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 0.535 ms | 0 - 2 MB | NPU
| FaceDetector | TFLITE | float | Qualcomm® SA8775P | 1.054 ms | 0 - 31 MB | NPU
| FaceDetector | TFLITE | float | Qualcomm® QCS9075 | 0.898 ms | 0 - 3 MB | NPU
| FaceDetector | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 0.767 ms | 0 - 38 MB | NPU
| FaceDetector | TFLITE | float | Qualcomm® SA7255P | 2.946 ms | 0 - 28 MB | NPU
| FaceDetector | TFLITE | float | Qualcomm® SA8295P | 1.101 ms | 0 - 24 MB | NPU
| FaceDetector | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.349 ms | 0 - 28 MB | NPU
| FaceDetector | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.28 ms | 0 - 32 MB | NPU
| FaceDetector | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.165 ms | 0 - 36 MB | NPU
| FaceDetector | TFLITE | w8a8 | Qualcomm® QCS6490 | 0.675 ms | 0 - 3 MB | NPU
| FaceDetector | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 0.683 ms | 0 - 27 MB | NPU
| FaceDetector | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.247 ms | 0 - 2 MB | NPU
| FaceDetector | TFLITE | w8a8 | Qualcomm® SA8775P | 1.569 ms | 0 - 27 MB | NPU
| FaceDetector | TFLITE | w8a8 | Qualcomm® QCS9075 | 0.373 ms | 0 - 2 MB | NPU
| FaceDetector | TFLITE | w8a8 | Qualcomm® QCM6690 | 1.257 ms | 0 - 25 MB | NPU
| FaceDetector | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.302 ms | 0 - 37 MB | NPU
| FaceDetector | TFLITE | w8a8 | Qualcomm® SA7255P | 0.683 ms | 0 - 27 MB | NPU
| FaceDetector | TFLITE | w8a8 | Qualcomm® SA8295P | 0.622 ms | 0 - 25 MB | NPU
| FaceDetector | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.151 ms | 0 - 30 MB | NPU
| FaceDetector | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.257 ms | 0 - 25 MB | NPU
| FaceDetector | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.131 ms | 0 - 30 MB | NPU
| FaceLandmarkDetector | ONNX | float | Snapdragon® X Elite | 0.382 ms | 0 - 0 MB | NPU
| FaceLandmarkDetector | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.205 ms | 0 - 34 MB | NPU
| FaceLandmarkDetector | ONNX | float | Qualcomm® QCS8550 (Proxy) | 0.282 ms | 0 - 17 MB | NPU
| FaceLandmarkDetector | ONNX | float | Qualcomm® QCS9075 | 0.427 ms | 0 - 3 MB | NPU
| FaceLandmarkDetector | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.188 ms | 0 - 22 MB | NPU
| FaceLandmarkDetector | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.149 ms | 0 - 27 MB | NPU
| FaceLandmarkDetector | ONNX | float | Snapdragon® X2 Elite | 0.152 ms | 0 - 0 MB | NPU
| FaceLandmarkDetector | ONNX | w8a8 | Snapdragon® X Elite | 0.298 ms | 0 - 0 MB | NPU
| FaceLandmarkDetector | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.154 ms | 0 - 36 MB | NPU
| FaceLandmarkDetector | ONNX | w8a8 | Qualcomm® QCS6490 | 3.735 ms | 5 - 8 MB | CPU
| FaceLandmarkDetector | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.202 ms | 0 - 3 MB | NPU
| FaceLandmarkDetector | ONNX | w8a8 | Qualcomm® QCS9075 | 0.347 ms | 0 - 3 MB | NPU
| FaceLandmarkDetector | ONNX | w8a8 | Qualcomm® QCM6690 | 3.769 ms | 0 - 7 MB | CPU
| FaceLandmarkDetector | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.133 ms | 0 - 27 MB | NPU
| FaceLandmarkDetector | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 2.462 ms | 0 - 8 MB | CPU
| FaceLandmarkDetector | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.126 ms | 0 - 30 MB | NPU
| FaceLandmarkDetector | ONNX | w8a8 | Snapdragon® X2 Elite | 0.131 ms | 0 - 0 MB | NPU
| FaceLandmarkDetector | QNN_DLC | float | Snapdragon® X Elite | 0.314 ms | 0 - 0 MB | NPU
| FaceLandmarkDetector | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.168 ms | 0 - 33 MB | NPU
| FaceLandmarkDetector | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 0.855 ms | 0 - 23 MB | NPU
| FaceLandmarkDetector | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 0.219 ms | 0 - 18 MB | NPU
| FaceLandmarkDetector | QNN_DLC | float | Qualcomm® SA8775P | 0.439 ms | 0 - 24 MB | NPU
| FaceLandmarkDetector | QNN_DLC | float | Qualcomm® QCS9075 | 0.331 ms | 0 - 2 MB | NPU
| FaceLandmarkDetector | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 0.325 ms | 0 - 37 MB | NPU
| FaceLandmarkDetector | QNN_DLC | float | Qualcomm® SA7255P | 0.855 ms | 0 - 23 MB | NPU
| FaceLandmarkDetector | QNN_DLC | float | Qualcomm® SA8295P | 0.598 ms | 0 - 23 MB | NPU
| FaceLandmarkDetector | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.141 ms | 0 - 22 MB | NPU
| FaceLandmarkDetector | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.124 ms | 0 - 26 MB | NPU
| FaceLandmarkDetector | QNN_DLC | float | Snapdragon® X2 Elite | 0.237 ms | 0 - 0 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.248 ms | 0 - 0 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.111 ms | 0 - 33 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 0.56 ms | 0 - 2 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 0.494 ms | 0 - 24 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.152 ms | 0 - 1 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.322 ms | 0 - 26 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 0.225 ms | 2 - 4 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 0.512 ms | 0 - 22 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.201 ms | 0 - 35 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Qualcomm® SA7255P | 0.494 ms | 0 - 24 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Qualcomm® SA8295P | 0.454 ms | 0 - 21 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.105 ms | 0 - 28 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.143 ms | 0 - 24 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.099 ms | 0 - 26 MB | NPU
| FaceLandmarkDetector | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.189 ms | 0 - 0 MB | NPU
| FaceLandmarkDetector | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.15 ms | 0 - 33 MB | NPU
| FaceLandmarkDetector | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 0.856 ms | 0 - 23 MB | NPU
| FaceLandmarkDetector | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 0.187 ms | 0 - 1 MB | NPU
| FaceLandmarkDetector | TFLITE | float | Qualcomm® SA8775P | 0.44 ms | 0 - 25 MB | NPU
| FaceLandmarkDetector | TFLITE | float | Qualcomm® QCS9075 | 0.313 ms | 0 - 4 MB | NPU
| FaceLandmarkDetector | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 0.313 ms | 0 - 38 MB | NPU
| FaceLandmarkDetector | TFLITE | float | Qualcomm® SA7255P | 0.856 ms | 0 - 23 MB | NPU
| FaceLandmarkDetector | TFLITE | float | Qualcomm® SA8295P | 0.572 ms | 0 - 24 MB | NPU
| FaceLandmarkDetector | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.132 ms | 0 - 27 MB | NPU
| FaceLandmarkDetector | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.108 ms | 0 - 26 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.102 ms | 0 - 32 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Qualcomm® QCS6490 | 0.476 ms | 0 - 3 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 0.508 ms | 0 - 23 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.152 ms | 0 - 2 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Qualcomm® SA8775P | 0.342 ms | 0 - 25 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Qualcomm® QCS9075 | 0.245 ms | 0 - 3 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Qualcomm® QCM6690 | 0.511 ms | 0 - 20 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.202 ms | 0 - 34 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Qualcomm® SA7255P | 0.508 ms | 0 - 23 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Qualcomm® SA8295P | 0.463 ms | 0 - 21 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.106 ms | 0 - 27 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.152 ms | 0 - 22 MB | NPU
| FaceLandmarkDetector | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.098 ms | 0 - 25 MB | NPU

## License
* The license for the original implementation of MediaPipe-Face-Detection can be found
  [here](https://github.com/zmurez/MediaPipePyTorch/blob/master/LICENSE).

## References
* [BlazeFace: Sub-millisecond Neural Face Detection on Mobile GPUs](https://arxiv.org/abs/1907.05047)
* [Source Model Implementation](https://github.com/zmurez/MediaPipePyTorch/)

## 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).