YOLO26-Pose: Optimized for Qualcomm Devices
Ultralytics YOLO26-Pose is a machine learning model that predicts human body keypoints and bounding boxes in an image.
This is based on the implementation of YOLO26-Pose 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
Due to licensing restrictions, we cannot distribute pre-exported model assets for this model. 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
See our repository for YOLO26-Pose on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.pose_estimation
Model Stats:
- Input resolution: 640x640
- Model checkpoint: YOLO26-N-Pose
- Model size (float): 11.4 MB
- Number of parameters: 2.97M
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| YOLO26-Pose | ONNX | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 3.84 ms | 6 - 181 MB | NPU |
| YOLO26-Pose | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.033 ms | 5 - 174 MB | NPU |
| YOLO26-Pose | ONNX | float | Snapdragon® X2 Elite | 3.882 ms | 9 - 9 MB | NPU |
| YOLO26-Pose | ONNX | float | Snapdragon® X Elite | 9.037 ms | 9 - 9 MB | NPU |
| YOLO26-Pose | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 6.147 ms | 8 - 212 MB | NPU |
| YOLO26-Pose | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 15.143 ms | 9 - 226 MB | NPU |
| YOLO26-Pose | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 9.438 ms | 6 - 10 MB | NPU |
| YOLO26-Pose | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 8.693 ms | 2 - 13 MB | NPU |
| YOLO26-Pose | ONNX | float | Qualcomm® QCS8450 | 15.143 ms | 9 - 226 MB | NPU |
| YOLO26-Pose | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 10.083 ms | 6 - 10 MB | NPU |
| YOLO26-Pose | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 9.037 ms | 9 - 9 MB | NPU |
| YOLO26-Pose | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 4.033 ms | 5 - 174 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 2.437 ms | 1 - 96 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 2.789 ms | 0 - 94 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Snapdragon® X2 Elite | 2.608 ms | 4 - 4 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Snapdragon® X Elite | 6.427 ms | 2 - 2 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 3.846 ms | 0 - 228 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 7.756 ms | 4 - 230 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 21.885 ms | 3 - 7 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 5.631 ms | 3 - 7 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 6.031 ms | 3 - 8 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Qualcomm® QCS8450 | 7.756 ms | 4 - 230 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 6.797 ms | 3 - 7 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 6.427 ms | 2 - 2 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 29.576 ms | 4 - 202 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 7.969 ms | 4 - 208 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 2.789 ms | 0 - 94 MB | NPU |
| YOLO26-Pose | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 7.969 ms | 4 - 208 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 2.613 ms | 2 - 174 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.133 ms | 0 - 154 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Snapdragon® X2 Elite | 3.339 ms | 5 - 5 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Snapdragon® X Elite | 6.269 ms | 5 - 5 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 4.102 ms | 5 - 193 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 10.08 ms | 3 - 206 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 6.748 ms | 5 - 14 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 5.691 ms | 5 - 6 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Qualcomm® SA8775P | 7.243 ms | 1 - 166 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Qualcomm® SA8650P | 7.243 ms | 1 - 166 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Qualcomm® SA8255P | 7.243 ms | 1 - 166 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Qualcomm® QCS8450 | 10.08 ms | 3 - 206 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 7.054 ms | 5 - 13 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 6.269 ms | 5 - 5 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 3.133 ms | 0 - 154 MB | NPU |
| YOLO26-Pose | QNN_DLC | float | Qualcomm® SA8295P | 10.436 ms | 0 - 180 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 2.147 ms | 2 - 192 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 2.474 ms | 0 - 178 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 2.587 ms | 2 - 2 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Snapdragon® X Elite | 5.671 ms | 2 - 2 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 3.385 ms | 2 - 210 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 4.859 ms | 1 - 7 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 5.097 ms | 2 - 4 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Qualcomm® SA8650P | 5.663 ms | 2 - 188 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Qualcomm® SA8255P | 5.663 ms | 2 - 188 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 5.512 ms | 4 - 10 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 5.671 ms | 2 - 2 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 22.056 ms | 2 - 181 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 5.32 ms | 2 - 183 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 2.474 ms | 0 - 178 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Qualcomm® SA7255P | 9.286 ms | 0 - 179 MB | NPU |
| YOLO26-Pose | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 5.32 ms | 2 - 183 MB | NPU |
| YOLO26-Pose | TFLITE | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 2.808 ms | 0 - 185 MB | NPU |
| YOLO26-Pose | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.807 ms | 0 - 171 MB | NPU |
| YOLO26-Pose | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 4.888 ms | 1 - 196 MB | NPU |
| YOLO26-Pose | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 12.79 ms | 0 - 216 MB | NPU |
| YOLO26-Pose | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 7.846 ms | 2 - 23 MB | NPU |
| YOLO26-Pose | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 7.251 ms | 2 - 5 MB | NPU |
| YOLO26-Pose | TFLITE | float | Qualcomm® SA8775P | 23.839 ms | 0 - 32 MB | GPU |
| YOLO26-Pose | TFLITE | float | Qualcomm® SA8650P | 23.839 ms | 0 - 32 MB | GPU |
| YOLO26-Pose | TFLITE | float | Qualcomm® SA8255P | 23.839 ms | 0 - 32 MB | GPU |
| YOLO26-Pose | TFLITE | float | Qualcomm® QCS8450 | 12.79 ms | 0 - 216 MB | NPU |
| YOLO26-Pose | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 8.52 ms | 2 - 22 MB | NPU |
| YOLO26-Pose | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 3.807 ms | 0 - 171 MB | NPU |
| YOLO26-Pose | TFLITE | float | Qualcomm® SA7255P | 18.207 ms | 2 - 175 MB | NPU |
| YOLO26-Pose | TFLITE | float | Qualcomm® SA8295P | 12.072 ms | 2 - 194 MB | NPU |
License
- The license for the original implementation of YOLO26-Pose can be found here.
References
Community
- Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
- For questions or feedback please reach out to us.
