Hugging Face's logo

qualcomm
/
Segformer-Base

Image Segmentation
PyTorch
real_time
android
Model card Files
xet
 Community

Segformer-Base: Optimized for Qualcomm Devices

Segformer Base is a machine learning model that predicts masks and classes of objects in an image.

This is based on the implementation of Segformer-Base found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up 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 2.45, ONNX Runtime 1.25.0 Download
ONNX w8a16 Universal QAIRT 2.45, ONNX Runtime 1.25.0 Download
ONNX w8a8 Universal QAIRT 2.45, ONNX Runtime 1.25.0 Download
QNN_DLC float Universal QAIRT 2.45 Download
QNN_DLC w8a16 Universal QAIRT 2.45 Download
QNN_DLC w8a8 Universal QAIRT 2.45 Download
TFLITE float Universal QAIRT 2.45 Download
TFLITE w8a8 Universal QAIRT 2.45 Download

For more device-specific assets and performance metrics, visit Segformer-Base on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

Use the Qualcomm® AI Hub Models 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 Segformer-Base on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

  • Model checkpoint: nvidia/segformer-b0-finetuned-ade-512-512
  • Input resolution: 512x512
  • Number of output classes: 150
  • Number of parameters: 3.75M
  • Model size (float): 14.4 MB
  • Model size (w8a16): 4.57 MB
  • Model size (w8a8): 3.90 MB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
Segformer-Base ONNX float Snapdragon® X2 Elite 13.579 ms 210 - 210 MB NPU
Segformer-Base ONNX float Snapdragon® X Elite 24.653 ms 146 - 146 MB NPU
Segformer-Base ONNX float Snapdragon® 8 Gen 3 Mobile 17.32 ms 10 - 216 MB NPU
Segformer-Base ONNX float Snapdragon® 8 Gen 1 Mobile 24.691 ms 23 - 233 MB NPU
Segformer-Base ONNX float Qualcomm® QCS8550 (Proxy) 23.503 ms 20 - 33 MB NPU
Segformer-Base ONNX float Qualcomm® QCS8450 24.691 ms 23 - 233 MB NPU
Segformer-Base ONNX float Snapdragon® 8 Elite Mobile 14.357 ms 23 - 209 MB NPU
Segformer-Base ONNX float Snapdragon® 8 Elite Gen 5 Mobile 13.253 ms 21 - 208 MB NPU
Segformer-Base ONNX float Qualcomm® QCS9075 25.591 ms 20 - 70 MB NPU
Segformer-Base ONNX float Qualcomm® QCS8750 14.357 ms 23 - 209 MB NPU
Segformer-Base ONNX float Qualcomm® QCS7181 24.653 ms 146 - 146 MB NPU
Segformer-Base ONNX w8a16 Snapdragon® X2 Elite 5.618 ms 211 - 211 MB NPU
Segformer-Base ONNX w8a16 Snapdragon® X Elite 13.574 ms 148 - 148 MB NPU
Segformer-Base ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 9.42 ms 12 - 235 MB NPU
Segformer-Base ONNX w8a16 Qualcomm® QCS8550 (Proxy) 13.382 ms 11 - 15 MB NPU
Segformer-Base ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 5.651 ms 14 - 215 MB NPU
Segformer-Base ONNX w8a16 Snapdragon® 8 Elite Mobile 7.247 ms 13 - 216 MB NPU
Segformer-Base ONNX w8a16 Qualcomm® QCS9075 18.538 ms 10 - 58 MB NPU
Segformer-Base ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 17.006 ms 13 - 233 MB NPU
Segformer-Base ONNX w8a16 Qualcomm® QCM6690 91.124 ms 4 - 230 MB NPU
Segformer-Base ONNX w8a16 Qualcomm® QCS7790 17.006 ms 13 - 233 MB NPU
Segformer-Base ONNX w8a16 Qualcomm® QCS8750 7.247 ms 13 - 216 MB NPU
Segformer-Base ONNX w8a16 Qualcomm® QCS7181 13.574 ms 148 - 148 MB NPU
Segformer-Base ONNX w8a8 Snapdragon® X2 Elite 1.941 ms 212 - 212 MB NPU
Segformer-Base ONNX w8a8 Snapdragon® X Elite 4.959 ms 148 - 148 MB NPU
Segformer-Base ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 3.221 ms 0 - 204 MB NPU
Segformer-Base ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 7.016 ms 6 - 212 MB NPU
Segformer-Base ONNX w8a8 Qualcomm® QCS6490 14.314 ms 5 - 51 MB NPU
Segformer-Base ONNX w8a8 Qualcomm® QCS8550 (Proxy) 4.767 ms 5 - 14 MB NPU
Segformer-Base ONNX w8a8 Qualcomm® QCS8450 7.016 ms 6 - 212 MB NPU
Segformer-Base ONNX w8a8 Qualcomm® QCS9075 5.622 ms 2 - 52 MB NPU
Segformer-Base ONNX w8a8 Qualcomm® QCM6690 31.491 ms 0 - 225 MB NPU
Segformer-Base ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 5.383 ms 8 - 201 MB NPU
Segformer-Base ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 1.913 ms 0 - 193 MB NPU
Segformer-Base ONNX w8a8 Snapdragon® 8 Elite Mobile 2.389 ms 0 - 191 MB NPU
Segformer-Base ONNX w8a8 Qualcomm® QCS7790 5.383 ms 8 - 201 MB NPU
Segformer-Base ONNX w8a8 Qualcomm® QCS8750 2.389 ms 0 - 191 MB NPU
Segformer-Base ONNX w8a8 Qualcomm® QCS7181 4.959 ms 148 - 148 MB NPU
Segformer-Base QNN_DLC float Snapdragon® X2 Elite 12.46 ms 3 - 3 MB NPU
Segformer-Base QNN_DLC float Snapdragon® X Elite 23.692 ms 3 - 3 MB NPU
Segformer-Base QNN_DLC float Snapdragon® 8 Gen 3 Mobile 16.334 ms 0 - 225 MB NPU
Segformer-Base QNN_DLC float Snapdragon® 8 Gen 1 Mobile 29.316 ms 0 - 223 MB NPU
Segformer-Base QNN_DLC float Qualcomm® QCS8275 48.87 ms 0 - 189 MB NPU
Segformer-Base QNN_DLC float Qualcomm® QCS8550 (Proxy) 22.625 ms 3 - 5 MB NPU
Segformer-Base QNN_DLC float Qualcomm® QCS8450 29.316 ms 0 - 223 MB NPU
Segformer-Base QNN_DLC float Snapdragon® 8 Elite Mobile 13.196 ms 3 - 200 MB NPU
Segformer-Base QNN_DLC float Qualcomm® SA7255P 48.87 ms 0 - 189 MB NPU
Segformer-Base QNN_DLC float Qualcomm® SA8295P 27.829 ms 0 - 192 MB NPU
Segformer-Base QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 13.696 ms 3 - 192 MB NPU
Segformer-Base QNN_DLC float Qualcomm® QCS9075 27.828 ms 3 - 17 MB NPU
Segformer-Base QNN_DLC float Qualcomm® QCS8750 13.196 ms 3 - 200 MB NPU
Segformer-Base QNN_DLC float Qualcomm® QCS7181 23.692 ms 3 - 3 MB NPU
Segformer-Base QNN_DLC w8a16 Snapdragon® X2 Elite 8.904 ms 2 - 2 MB NPU
Segformer-Base QNN_DLC w8a16 Snapdragon® X Elite 19.637 ms 2 - 2 MB NPU
Segformer-Base QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 15.333 ms 2 - 240 MB NPU
Segformer-Base QNN_DLC w8a16 Qualcomm® QCS8275 32.41 ms 2 - 204 MB NPU
Segformer-Base QNN_DLC w8a16 Qualcomm® QCS8550 (Proxy) 20.536 ms 2 - 3 MB NPU
Segformer-Base QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 9.732 ms 2 - 216 MB NPU
Segformer-Base QNN_DLC w8a16 Snapdragon® 8 Elite Mobile 11.302 ms 1 - 206 MB NPU
Segformer-Base QNN_DLC w8a16 Qualcomm® QCS9075 33.911 ms 1 - 9 MB NPU
Segformer-Base QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 34.451 ms 2 - 232 MB NPU
Segformer-Base QNN_DLC w8a16 Qualcomm® QCM6690 129.787 ms 2 - 254 MB NPU
Segformer-Base QNN_DLC w8a16 Qualcomm® SA7255P 32.41 ms 2 - 204 MB NPU
Segformer-Base QNN_DLC w8a16 Qualcomm® QCS7790 34.451 ms 2 - 232 MB NPU
Segformer-Base QNN_DLC w8a16 Qualcomm® QCS8750 11.302 ms 1 - 206 MB NPU
Segformer-Base QNN_DLC w8a16 Qualcomm® QCS7181 19.637 ms 2 - 2 MB NPU
Segformer-Base QNN_DLC w8a8 Snapdragon® X2 Elite 2.541 ms 1 - 1 MB NPU
Segformer-Base QNN_DLC w8a8 Snapdragon® X Elite 6.363 ms 1 - 1 MB NPU
Segformer-Base QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 3.955 ms 1 - 201 MB NPU
Segformer-Base QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 8.641 ms 0 - 196 MB NPU
Segformer-Base QNN_DLC w8a8 Qualcomm® QCS6490 15.677 ms 1 - 5 MB NPU
Segformer-Base QNN_DLC w8a8 Qualcomm® QCS8275 10.865 ms 1 - 174 MB NPU
Segformer-Base QNN_DLC w8a8 Qualcomm® QCS8550 (Proxy) 5.787 ms 1 - 10 MB NPU
Segformer-Base QNN_DLC w8a8 Qualcomm® QCS8450 8.641 ms 0 - 196 MB NPU
Segformer-Base QNN_DLC w8a8 Qualcomm® QCS9075 6.706 ms 0 - 5 MB NPU
Segformer-Base QNN_DLC w8a8 Qualcomm® SA7255P 10.865 ms 1 - 174 MB NPU
Segformer-Base QNN_DLC w8a8 Qualcomm® QCM6690 35.336 ms 1 - 188 MB NPU
Segformer-Base QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 6.84 ms 1 - 182 MB NPU
Segformer-Base QNN_DLC w8a8 Qualcomm® SA8295P 7.765 ms 0 - 172 MB NPU
Segformer-Base QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 2.243 ms 1 - 184 MB NPU
Segformer-Base QNN_DLC w8a8 Snapdragon® 8 Elite Mobile 2.841 ms 0 - 178 MB NPU
Segformer-Base QNN_DLC w8a8 Qualcomm® QCS7790 6.84 ms 1 - 182 MB NPU
Segformer-Base QNN_DLC w8a8 Qualcomm® QCS8750 2.841 ms 0 - 178 MB NPU
Segformer-Base QNN_DLC w8a8 Qualcomm® QCS7181 6.363 ms 1 - 1 MB NPU
Segformer-Base TFLITE float Snapdragon® 8 Gen 3 Mobile 16.521 ms 9 - 241 MB NPU
Segformer-Base TFLITE float Snapdragon® 8 Gen 1 Mobile 29.561 ms 9 - 234 MB NPU
Segformer-Base TFLITE float Qualcomm® QCS8275 48.908 ms 10 - 200 MB NPU
Segformer-Base TFLITE float Qualcomm® QCS8550 (Proxy) 22.647 ms 9 - 12 MB NPU
Segformer-Base TFLITE float Qualcomm® SA8775P 99.841 ms 10 - 40 MB GPU
Segformer-Base TFLITE float Qualcomm® SA8650P 99.841 ms 10 - 40 MB GPU
Segformer-Base TFLITE float Qualcomm® SA8255P 99.841 ms 10 - 40 MB GPU
Segformer-Base TFLITE float Qualcomm® QCS8450 29.561 ms 9 - 234 MB NPU
Segformer-Base TFLITE float Snapdragon® 8 Elite Mobile 13.299 ms 9 - 205 MB NPU
Segformer-Base TFLITE float Qualcomm® SA7255P 48.908 ms 10 - 200 MB NPU
Segformer-Base TFLITE float Qualcomm® SA8295P 27.821 ms 9 - 203 MB NPU
Segformer-Base TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 13.644 ms 8 - 200 MB NPU
Segformer-Base TFLITE float Qualcomm® QCS9075 27.587 ms 8 - 30 MB NPU
Segformer-Base TFLITE float Qualcomm® QCS8750 13.299 ms 9 - 205 MB NPU
Segformer-Base TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 7.108 ms 2 - 212 MB NPU
Segformer-Base TFLITE w8a8 Snapdragon® 8 Gen 1 Mobile 12.13 ms 0 - 208 MB NPU
Segformer-Base TFLITE w8a8 Qualcomm® QCS6490 110.956 ms 15 - 50 MB NPU
Segformer-Base TFLITE w8a8 Qualcomm® QCS8275 18.289 ms 2 - 179 MB NPU
Segformer-Base TFLITE w8a8 Qualcomm® QCS8550 (Proxy) 10.124 ms 2 - 5 MB NPU
Segformer-Base TFLITE w8a8 Qualcomm® SA8775P 106.739 ms 15 - 47 MB GPU
Segformer-Base TFLITE w8a8 Qualcomm® SA8650P 106.739 ms 15 - 47 MB GPU
Segformer-Base TFLITE w8a8 Qualcomm® SA8255P 106.739 ms 15 - 47 MB GPU
Segformer-Base TFLITE w8a8 Qualcomm® QCS8450 12.13 ms 0 - 208 MB NPU
Segformer-Base TFLITE w8a8 Qualcomm® QCS9075 10.652 ms 2 - 12 MB NPU
Segformer-Base TFLITE w8a8 Qualcomm® SA7255P 18.289 ms 2 - 179 MB NPU
Segformer-Base TFLITE w8a8 Qualcomm® QCM6690 95.14 ms 13 - 185 MB NPU
Segformer-Base TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 38.935 ms 15 - 73 MB NPU
Segformer-Base TFLITE w8a8 Qualcomm® SA8295P 12.874 ms 2 - 184 MB NPU
Segformer-Base TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 4.547 ms 2 - 182 MB NPU
Segformer-Base TFLITE w8a8 Snapdragon® 8 Elite Mobile 5.723 ms 0 - 177 MB NPU
Segformer-Base TFLITE w8a8 Qualcomm® QCS7790 38.935 ms 15 - 73 MB NPU
Segformer-Base TFLITE w8a8 Qualcomm® QCS8750 5.723 ms 0 - 177 MB NPU

License

  • The license for the original implementation of Segformer-Base can be found here.

References

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
Image Segmentation
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Paper for qualcomm/Segformer-Base