ControlNet-Canny: Optimized for Qualcomm Devices

On-device, high-resolution image synthesis from text and image prompts. ControlNet guides Stable-diffusion with provided input image to generate accurate images from given input prompt.

This is based on the implementation of ControlNet-Canny 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
PRECOMPILED_QNN_ONNX w8a16 Snapdragon® X2 Elite QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX w8a16 Snapdragon® X Elite QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Gen 3 Mobile QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 7 Gen 4 Mobile QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Elite Mobile QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile QAIRT 2.45, ONNX Runtime 1.27.1 Download
QNN_CONTEXT_BINARY w8a16 Snapdragon® X2 Elite QAIRT 2.45 Download
QNN_CONTEXT_BINARY w8a16 Snapdragon® X Elite QAIRT 2.45 Download
QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Gen 3 Mobile QAIRT 2.45 Download
QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) QAIRT 2.45 Download
QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8775P QAIRT 2.45 Download
QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-6690 QAIRT 2.45 Download
QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ IQ-9075 QAIRT 2.45 Download
QNN_CONTEXT_BINARY w8a16 Qualcomm® SA7255P QAIRT 2.45 Download
QNN_CONTEXT_BINARY w8a16 Snapdragon® 7 Gen 4 Mobile QAIRT 2.45 Download
QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Elite Mobile QAIRT 2.45 Download
QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Elite Gen 5 Mobile QAIRT 2.45 Download

For more device-specific assets and performance metrics, visit ControlNet-Canny 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 ControlNet-Canny on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.image_generation

Model Stats:

  • Input: Text prompt and input image as a reference
  • Conditioning Input: Canny-Edge
  • Text Encoder Number of parameters: 340M
  • UNet Number of parameters: 865M
  • VAE Decoder Number of parameters: 83M
  • ControlNet Number of parameters: 361M
  • Model size: 1.4GB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
controlnet PRECOMPILED_QNN_ONNX w8a16 Snapdragon® X2 Elite 29.153 ms 34 - 34 MB NPU
controlnet PRECOMPILED_QNN_ONNX w8a16 Snapdragon® X Elite 70.852 ms 352 - 352 MB NPU
controlnet PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 59.536 ms 33 - 44 MB NPU
controlnet PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 75.288 ms 30 - 33 MB NPU
controlnet PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 70.934 ms 31 - 36 MB NPU
controlnet PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 70.852 ms 352 - 352 MB NPU
controlnet PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 1108.537 ms 2 - 14 MB NPU
controlnet PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 120.573 ms 32 - 42 MB NPU
controlnet PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 44.199 ms 34 - 45 MB NPU
controlnet PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Elite Mobile 44.199 ms 34 - 45 MB NPU
controlnet PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 28.305 ms 32 - 41 MB NPU
controlnet PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 120.573 ms 32 - 42 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Snapdragon® X2 Elite 29.535 ms 2 - 2 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Snapdragon® X Elite 71.04 ms 2 - 2 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Gen 3 Mobile 57.433 ms 2 - 10 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ QCS8275 137.881 ms 2 - 11 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 74.967 ms 2 - 4 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8775P 70.403 ms 2 - 11 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8650P 70.403 ms 2 - 11 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8255P 70.403 ms 2 - 11 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ IQ-9075 70.558 ms 0 - 16 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ IQ-X7181 71.04 ms 2 - 2 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-6690 1026.916 ms 4 - 10 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-7790 120.022 ms 2 - 8 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-8750 43.941 ms 2 - 11 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Qualcomm® SA7255P 137.881 ms 2 - 11 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Elite Mobile 43.941 ms 2 - 11 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Elite Gen 5 Mobile 28.405 ms 2 - 10 MB NPU
controlnet QNN_CONTEXT_BINARY w8a16 Snapdragon® 7 Gen 4 Mobile 120.022 ms 2 - 8 MB NPU
text_encoder PRECOMPILED_QNN_ONNX w8a16 Snapdragon® X2 Elite 1.517 ms 1 - 1 MB NPU
text_encoder PRECOMPILED_QNN_ONNX w8a16 Snapdragon® X Elite 3.436 ms 157 - 157 MB NPU
text_encoder PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 2.315 ms 0 - 12 MB NPU
text_encoder PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 3.246 ms 0 - 162 MB NPU
text_encoder PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 3.34 ms 0 - 3 MB NPU
text_encoder PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 3.436 ms 157 - 157 MB NPU
text_encoder PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 22.058 ms 0 - 10 MB NPU
text_encoder PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 3.689 ms 0 - 6 MB NPU
text_encoder PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 1.749 ms 0 - 7 MB NPU
text_encoder PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Elite Mobile 1.749 ms 0 - 7 MB NPU
text_encoder PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 1.543 ms 0 - 7 MB NPU
text_encoder PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 3.689 ms 0 - 6 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Snapdragon® X2 Elite 1.725 ms 0 - 0 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Snapdragon® X Elite 3.606 ms 1 - 1 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Gen 3 Mobile 2.296 ms 0 - 7 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ QCS8275 7.488 ms 0 - 9 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 3.249 ms 0 - 1 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8775P 3.506 ms 0 - 10 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8650P 3.506 ms 0 - 10 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8255P 3.506 ms 0 - 10 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ IQ-9075 4.498 ms 0 - 2 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ IQ-X7181 3.606 ms 1 - 1 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-6690 22.948 ms 0 - 6 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-7790 3.687 ms 0 - 6 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-8750 1.757 ms 0 - 14 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Qualcomm® SA7255P 7.488 ms 0 - 9 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Elite Mobile 1.757 ms 0 - 14 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Elite Gen 5 Mobile 1.544 ms 0 - 9 MB NPU
text_encoder QNN_CONTEXT_BINARY w8a16 Snapdragon® 7 Gen 4 Mobile 3.687 ms 0 - 6 MB NPU
unet PRECOMPILED_QNN_ONNX w8a16 Snapdragon® X2 Elite 46.522 ms 13 - 13 MB NPU
unet PRECOMPILED_QNN_ONNX w8a16 Snapdragon® X Elite 113.927 ms 829 - 829 MB NPU
unet PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 80.831 ms 13 - 25 MB NPU
unet PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 113.202 ms 0 - 883 MB NPU
unet PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 106.676 ms 12 - 28 MB NPU
unet PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 113.927 ms 829 - 829 MB NPU
unet PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 1942.342 ms 14 - 24 MB NPU
unet PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 179.92 ms 13 - 24 MB NPU
unet PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 64.095 ms 6 - 17 MB NPU
unet PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Elite Mobile 64.095 ms 6 - 17 MB NPU
unet PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 50.382 ms 6 - 17 MB NPU
unet PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 179.92 ms 13 - 24 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Snapdragon® X2 Elite 47.081 ms 13 - 13 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Snapdragon® X Elite 114.222 ms 13 - 13 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Gen 3 Mobile 81.038 ms 13 - 20 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ QCS8275 266.464 ms 10 - 19 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 112.56 ms 13 - 15 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8775P 107.105 ms 6 - 16 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8650P 107.105 ms 6 - 16 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8255P 107.105 ms 6 - 16 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ IQ-9075 106.489 ms 11 - 25 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ IQ-X7181 114.222 ms 13 - 13 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-6690 1981.817 ms 0 - 7 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-7790 180.344 ms 11 - 18 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-8750 61.939 ms 6 - 15 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Qualcomm® SA7255P 266.464 ms 10 - 19 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Elite Mobile 61.939 ms 6 - 15 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Elite Gen 5 Mobile 46.201 ms 6 - 14 MB NPU
unet QNN_CONTEXT_BINARY w8a16 Snapdragon® 7 Gen 4 Mobile 180.344 ms 11 - 18 MB NPU
vae PRECOMPILED_QNN_ONNX w8a16 Snapdragon® X2 Elite 127.148 ms 4 - 4 MB NPU
vae PRECOMPILED_QNN_ONNX w8a16 Snapdragon® X Elite 258.713 ms 62 - 62 MB NPU
vae PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 197.319 ms 3 - 11 MB NPU
vae PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 267.188 ms 4 - 5 MB NPU
vae PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 240.505 ms 4 - 6 MB NPU
vae PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 258.713 ms 62 - 62 MB NPU
vae PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 6298.039 ms 0 - 8 MB NPU
vae PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 504.636 ms 2 - 8 MB NPU
vae PRECOMPILED_QNN_ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 173.077 ms 4 - 15 MB NPU
vae PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Elite Mobile 173.077 ms 4 - 15 MB NPU
vae PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 120.184 ms 3 - 10 MB NPU
vae PRECOMPILED_QNN_ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 504.636 ms 2 - 8 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Snapdragon® X2 Elite 126.912 ms 0 - 0 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Snapdragon® X Elite 258.849 ms 0 - 0 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Gen 3 Mobile 198.257 ms 3 - 10 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ QCS8275 711.699 ms 0 - 10 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 260.739 ms 0 - 2 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8775P 243.863 ms 0 - 10 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8650P 243.863 ms 0 - 10 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Qualcomm® SA8255P 243.863 ms 0 - 10 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ IQ-9075 240.43 ms 0 - 3 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ IQ-X7181 258.849 ms 0 - 0 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-6690 6359.803 ms 2 - 9 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-7790 503.632 ms 0 - 6 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Qualcomm® Dragonwing™ Q-8750 171.923 ms 0 - 13 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Qualcomm® SA7255P 711.699 ms 0 - 10 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Elite Mobile 171.923 ms 0 - 13 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Snapdragon® 8 Elite Gen 5 Mobile 120.071 ms 0 - 9 MB NPU
vae QNN_CONTEXT_BINARY w8a16 Snapdragon® 7 Gen 4 Mobile 503.632 ms 0 - 6 MB NPU

License

  • The license for the original implementation of ControlNet-Canny can be found here.

References

Community

Usage and Limitations

This model may not be used for or in connection with any of the following applications:

  • Accessing essential private and public services and benefits;
  • Administration of justice and democratic processes;
  • Assessing or recognizing the emotional state of a person;
  • Biometric and biometrics-based systems, including categorization of persons based on sensitive characteristics;
  • Education and vocational training;
  • Employment and workers management;
  • Exploitation of the vulnerabilities of persons resulting in harmful behavior;
  • General purpose social scoring;
  • Law enforcement;
  • Management and operation of critical infrastructure;
  • Migration, asylum and border control management;
  • Predictive policing;
  • Real-time remote biometric identification in public spaces;
  • Recommender systems of social media platforms;
  • Scraping of facial images (from the internet or otherwise); and/or
  • Subliminal manipulation
Downloads last month

-

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

Paper for qualcomm/ControlNet-Canny