v0.60.0
Browse filesSee https://github.com/qualcomm/ai-hub-models/releases/v0.60.0 for changelog.
README.md
CHANGED
|
@@ -15,18 +15,18 @@ pipeline_tag: object-detection
|
|
| 15 |
Ultralytics YOLOv10 is a machine learning model that predicts bounding boxes and classes of objects in an image.
|
| 16 |
|
| 17 |
This is based on the implementation of YOLOv10-Detection found [here](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/detect).
|
| 18 |
-
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/v0.
|
| 19 |
|
| 20 |
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.
|
| 21 |
|
| 22 |
## Getting Started
|
| 23 |
Due to licensing restrictions, we cannot distribute pre-exported model assets for this model.
|
| 24 |
-
Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.
|
| 25 |
- Custom weights (e.g., fine-tuned checkpoints)
|
| 26 |
- Custom input shapes
|
| 27 |
- Target device and runtime configurations
|
| 28 |
|
| 29 |
-
See our repository for [YOLOv10-Detection on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.
|
| 30 |
|
| 31 |
|
| 32 |
## Model Details
|
|
@@ -34,93 +34,101 @@ See our repository for [YOLOv10-Detection on GitHub](https://github.com/qualcomm
|
|
| 34 |
**Model Type:** Model_use_case.object_detection
|
| 35 |
|
| 36 |
**Model Stats:**
|
| 37 |
-
- Model checkpoint: YOLOv10-N
|
| 38 |
- Input resolution: 640x640
|
| 39 |
-
-
|
| 40 |
- Model size (float): 8.95 MB
|
| 41 |
-
- Model size (w8a8): 2.55 MB
|
| 42 |
- Model size (w8a16): 3.04 MB
|
|
|
|
|
|
|
| 43 |
|
| 44 |
## Performance Summary
|
| 45 |
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|
| 46 |
|---|---|---|---|---|---|---
|
| 47 |
-
| YOLOv10-Detection | ONNX | float | Snapdragon® X2 Elite | 3.
|
| 48 |
-
| YOLOv10-Detection | ONNX | float | Snapdragon® X Elite | 6.
|
| 49 |
-
| YOLOv10-Detection | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 3.
|
| 50 |
-
| YOLOv10-Detection | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 14.
|
| 51 |
-
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™
|
| 52 |
-
| YOLOv10-Detection | ONNX | float | Qualcomm®
|
| 53 |
-
| YOLOv10-Detection | ONNX | float | Qualcomm®
|
| 54 |
-
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™ IQ-
|
| 55 |
-
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™
|
| 56 |
-
| YOLOv10-Detection | ONNX | float |
|
| 57 |
-
| YOLOv10-Detection | ONNX | float | Snapdragon® 8 Elite
|
| 58 |
-
| YOLOv10-Detection | ONNX |
|
| 59 |
-
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon®
|
| 60 |
-
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon®
|
| 61 |
-
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® 8 Gen
|
| 62 |
-
| YOLOv10-Detection | ONNX | w8a16 |
|
| 63 |
-
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™
|
| 64 |
-
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm®
|
| 65 |
-
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™
|
| 66 |
-
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm®
|
| 67 |
-
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™
|
| 68 |
-
| YOLOv10-Detection | ONNX | w8a16 |
|
| 69 |
-
| YOLOv10-Detection | ONNX | w8a16 |
|
| 70 |
-
| YOLOv10-Detection | ONNX |
|
| 71 |
-
| YOLOv10-Detection | ONNX |
|
| 72 |
-
| YOLOv10-Detection | ONNX |
|
| 73 |
-
| YOLOv10-Detection | ONNX |
|
| 74 |
-
| YOLOv10-Detection | ONNX |
|
| 75 |
-
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® 8
|
| 76 |
-
| YOLOv10-Detection |
|
| 77 |
-
| YOLOv10-Detection |
|
| 78 |
-
| YOLOv10-Detection |
|
| 79 |
-
| YOLOv10-Detection |
|
| 80 |
-
| YOLOv10-Detection |
|
| 81 |
-
| YOLOv10-Detection |
|
| 82 |
-
| YOLOv10-Detection | QNN_DLC | float |
|
| 83 |
-
| YOLOv10-Detection | QNN_DLC | float |
|
| 84 |
-
| YOLOv10-Detection | QNN_DLC | float |
|
| 85 |
-
| YOLOv10-Detection | QNN_DLC | float |
|
| 86 |
-
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-
|
| 87 |
-
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-
|
| 88 |
-
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™
|
| 89 |
-
| YOLOv10-Detection | QNN_DLC | float | Qualcomm®
|
| 90 |
-
| YOLOv10-Detection | QNN_DLC | float | Qualcomm®
|
| 91 |
-
| YOLOv10-Detection | QNN_DLC | float |
|
| 92 |
-
| YOLOv10-Detection | QNN_DLC | float |
|
| 93 |
-
| YOLOv10-Detection | QNN_DLC |
|
| 94 |
-
| YOLOv10-Detection | QNN_DLC |
|
| 95 |
-
| YOLOv10-Detection | QNN_DLC |
|
| 96 |
-
| YOLOv10-Detection | QNN_DLC |
|
| 97 |
-
| YOLOv10-Detection | QNN_DLC |
|
| 98 |
-
| YOLOv10-Detection | QNN_DLC |
|
| 99 |
-
| YOLOv10-Detection | QNN_DLC |
|
| 100 |
-
| YOLOv10-Detection | QNN_DLC | w8a16 |
|
| 101 |
-
| YOLOv10-Detection | QNN_DLC | w8a16 |
|
| 102 |
-
| YOLOv10-Detection | QNN_DLC | w8a16 |
|
| 103 |
-
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™
|
| 104 |
-
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™
|
| 105 |
-
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™
|
| 106 |
-
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm®
|
| 107 |
-
| YOLOv10-Detection | QNN_DLC | w8a16 |
|
| 108 |
-
| YOLOv10-Detection | QNN_DLC | w8a16 |
|
| 109 |
-
| YOLOv10-Detection | QNN_DLC | w8a16 |
|
| 110 |
-
| YOLOv10-Detection |
|
| 111 |
-
| YOLOv10-Detection |
|
| 112 |
-
| YOLOv10-Detection |
|
| 113 |
-
| YOLOv10-Detection |
|
| 114 |
-
| YOLOv10-Detection |
|
| 115 |
-
| YOLOv10-Detection |
|
| 116 |
-
| YOLOv10-Detection |
|
| 117 |
-
| YOLOv10-Detection |
|
| 118 |
-
| YOLOv10-Detection | TFLITE | float |
|
| 119 |
-
| YOLOv10-Detection | TFLITE | float |
|
| 120 |
-
| YOLOv10-Detection | TFLITE | float | Qualcomm®
|
| 121 |
-
| YOLOv10-Detection | TFLITE | float | Qualcomm®
|
| 122 |
-
| YOLOv10-Detection | TFLITE | float |
|
| 123 |
-
| YOLOv10-Detection | TFLITE | float |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 124 |
|
| 125 |
## License
|
| 126 |
* The license for the original implementation of YOLOv10-Detection can be found
|
|
|
|
| 15 |
Ultralytics YOLOv10 is a machine learning model that predicts bounding boxes and classes of objects in an image.
|
| 16 |
|
| 17 |
This is based on the implementation of YOLOv10-Detection found [here](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/detect).
|
| 18 |
+
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/v0.60.0/src/qai_hub_models/models/yolov10_det) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
|
| 19 |
|
| 20 |
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.
|
| 21 |
|
| 22 |
## Getting Started
|
| 23 |
Due to licensing restrictions, we cannot distribute pre-exported model assets for this model.
|
| 24 |
+
Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/yolov10_det) Python library to compile and export the model with your own:
|
| 25 |
- Custom weights (e.g., fine-tuned checkpoints)
|
| 26 |
- Custom input shapes
|
| 27 |
- Target device and runtime configurations
|
| 28 |
|
| 29 |
+
See our repository for [YOLOv10-Detection on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/yolov10_det) for usage instructions.
|
| 30 |
|
| 31 |
|
| 32 |
## Model Details
|
|
|
|
| 34 |
**Model Type:** Model_use_case.object_detection
|
| 35 |
|
| 36 |
**Model Stats:**
|
|
|
|
| 37 |
- Input resolution: 640x640
|
| 38 |
+
- Model checkpoint: YOLOv10-N
|
| 39 |
- Model size (float): 8.95 MB
|
|
|
|
| 40 |
- Model size (w8a16): 3.04 MB
|
| 41 |
+
- Model size (w8a8): 2.55 MB
|
| 42 |
+
- Number of parameters: 2.33M
|
| 43 |
|
| 44 |
## Performance Summary
|
| 45 |
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|
| 46 |
|---|---|---|---|---|---|---
|
| 47 |
+
| YOLOv10-Detection | ONNX | float | Snapdragon® X2 Elite | 3.35 ms | 5 - 5 MB | NPU
|
| 48 |
+
| YOLOv10-Detection | ONNX | float | Snapdragon® X Elite | 6.738 ms | 5 - 5 MB | NPU
|
| 49 |
+
| YOLOv10-Detection | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 3.923 ms | 5 - 245 MB | NPU
|
| 50 |
+
| YOLOv10-Detection | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 14.875 ms | 5 - 215 MB | NPU
|
| 51 |
+
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 6.694 ms | 5 - 8 MB | NPU
|
| 52 |
+
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 6.342 ms | 5 - 9 MB | NPU
|
| 53 |
+
| YOLOv10-Detection | ONNX | float | Qualcomm® QCS8450 | 14.875 ms | 5 - 215 MB | NPU
|
| 54 |
+
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 7.609 ms | 5 - 7 MB | NPU
|
| 55 |
+
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 6.738 ms | 5 - 5 MB | NPU
|
| 56 |
+
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 4.386 ms | 1 - 201 MB | NPU
|
| 57 |
+
| YOLOv10-Detection | ONNX | float | Snapdragon® 8 Elite Mobile | 4.386 ms | 1 - 201 MB | NPU
|
| 58 |
+
| YOLOv10-Detection | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.026 ms | 1 - 219 MB | NPU
|
| 59 |
+
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® X2 Elite | 2.424 ms | 2 - 2 MB | NPU
|
| 60 |
+
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® X Elite | 5.684 ms | 2 - 2 MB | NPU
|
| 61 |
+
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 3.351 ms | 0 - 97 MB | NPU
|
| 62 |
+
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 6.68 ms | 3 - 101 MB | NPU
|
| 63 |
+
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 16.669 ms | 2 - 5 MB | NPU
|
| 64 |
+
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 4.806 ms | 2 - 6 MB | NPU
|
| 65 |
+
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 5.395 ms | 2 - 6 MB | NPU
|
| 66 |
+
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® QCS8450 | 6.68 ms | 3 - 101 MB | NPU
|
| 67 |
+
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 5.808 ms | 2 - 5 MB | NPU
|
| 68 |
+
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 5.684 ms | 2 - 2 MB | NPU
|
| 69 |
+
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 24.947 ms | 3 - 196 MB | NPU
|
| 70 |
+
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 6.66 ms | 3 - 199 MB | NPU
|
| 71 |
+
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 2.525 ms | 1 - 79 MB | NPU
|
| 72 |
+
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 2.525 ms | 1 - 79 MB | NPU
|
| 73 |
+
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 2.283 ms | 0 - 82 MB | NPU
|
| 74 |
+
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 6.66 ms | 3 - 199 MB | NPU
|
| 75 |
+
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 7.921 ms | 9 - 383 MB | NPU
|
| 76 |
+
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Qualcomm® Dragonwing™ IQ-8275 | 11.622 ms | 8 - 12 MB | NPU
|
| 77 |
+
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.198 ms | 8 - 12 MB | NPU
|
| 78 |
+
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Qualcomm® Dragonwing™ IQ-9075 | 12.222 ms | 8 - 11 MB | NPU
|
| 79 |
+
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Qualcomm® Dragonwing™ Q-8750 | 6.05 ms | 7 - 357 MB | NPU
|
| 80 |
+
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Elite Mobile | 6.05 ms | 7 - 357 MB | NPU
|
| 81 |
+
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 6.964 ms | 6 - 364 MB | NPU
|
| 82 |
+
| YOLOv10-Detection | QNN_DLC | float | Snapdragon® X2 Elite | 2.67 ms | 5 - 5 MB | NPU
|
| 83 |
+
| YOLOv10-Detection | QNN_DLC | float | Snapdragon® X Elite | 4.656 ms | 5 - 5 MB | NPU
|
| 84 |
+
| YOLOv10-Detection | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 3.035 ms | 0 - 210 MB | NPU
|
| 85 |
+
| YOLOv10-Detection | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 8.533 ms | 5 - 202 MB | NPU
|
| 86 |
+
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 5.175 ms | 5 - 12 MB | NPU
|
| 87 |
+
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 13.26 ms | 2 - 182 MB | NPU
|
| 88 |
+
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.144 ms | 5 - 7 MB | NPU
|
| 89 |
+
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® SA8775P | 5.624 ms | 0 - 186 MB | NPU
|
| 90 |
+
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® SA8650P | 5.624 ms | 0 - 186 MB | NPU
|
| 91 |
+
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® SA8255P | 5.624 ms | 0 - 186 MB | NPU
|
| 92 |
+
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® QCS8450 | 8.533 ms | 5 - 202 MB | NPU
|
| 93 |
+
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 5.382 ms | 5 - 11 MB | NPU
|
| 94 |
+
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 4.656 ms | 5 - 5 MB | NPU
|
| 95 |
+
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 2.269 ms | 5 - 191 MB | NPU
|
| 96 |
+
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® SA7255P | 13.26 ms | 2 - 182 MB | NPU
|
| 97 |
+
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® SA8295P | 8.923 ms | 0 - 169 MB | NPU
|
| 98 |
+
| YOLOv10-Detection | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 2.269 ms | 5 - 191 MB | NPU
|
| 99 |
+
| YOLOv10-Detection | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.988 ms | 5 - 199 MB | NPU
|
| 100 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 2.23 ms | 2 - 2 MB | NPU
|
| 101 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Snapdragon® X Elite | 4.619 ms | 2 - 2 MB | NPU
|
| 102 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 2.913 ms | 0 - 196 MB | NPU
|
| 103 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 4.12 ms | 1 - 6 MB | NPU
|
| 104 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 7.975 ms | 0 - 173 MB | NPU
|
| 105 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.211 ms | 2 - 105 MB | NPU
|
| 106 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® SA8775P | 4.828 ms | 1 - 175 MB | NPU
|
| 107 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® SA8650P | 4.828 ms | 1 - 175 MB | NPU
|
| 108 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® SA8255P | 4.828 ms | 1 - 175 MB | NPU
|
| 109 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 4.627 ms | 2 - 6 MB | NPU
|
| 110 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 4.619 ms | 2 - 2 MB | NPU
|
| 111 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 20.139 ms | 2 - 180 MB | NPU
|
| 112 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 4.444 ms | 2 - 180 MB | NPU
|
| 113 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 2.021 ms | 0 - 175 MB | NPU
|
| 114 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® SA7255P | 7.975 ms | 0 - 173 MB | NPU
|
| 115 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 2.021 ms | 0 - 175 MB | NPU
|
| 116 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 1.754 ms | 2 - 186 MB | NPU
|
| 117 |
+
| YOLOv10-Detection | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 4.444 ms | 2 - 180 MB | NPU
|
| 118 |
+
| YOLOv10-Detection | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 2.726 ms | 0 - 110 MB | NPU
|
| 119 |
+
| YOLOv10-Detection | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 7.443 ms | 0 - 85 MB | NPU
|
| 120 |
+
| YOLOv10-Detection | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 4.856 ms | 0 - 14 MB | NPU
|
| 121 |
+
| YOLOv10-Detection | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 12.756 ms | 0 - 79 MB | NPU
|
| 122 |
+
| YOLOv10-Detection | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 3.759 ms | 0 - 3 MB | NPU
|
| 123 |
+
| YOLOv10-Detection | TFLITE | float | Qualcomm® SA8775P | 5.206 ms | 0 - 84 MB | NPU
|
| 124 |
+
| YOLOv10-Detection | TFLITE | float | Qualcomm® SA8650P | 5.206 ms | 0 - 84 MB | NPU
|
| 125 |
+
| YOLOv10-Detection | TFLITE | float | Qualcomm® SA8255P | 5.206 ms | 0 - 84 MB | NPU
|
| 126 |
+
| YOLOv10-Detection | TFLITE | float | Qualcomm® QCS8450 | 7.443 ms | 0 - 85 MB | NPU
|
| 127 |
+
| YOLOv10-Detection | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 5.992 ms | 0 - 13 MB | NPU
|
| 128 |
+
| YOLOv10-Detection | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 2.042 ms | 0 - 80 MB | NPU
|
| 129 |
+
| YOLOv10-Detection | TFLITE | float | Qualcomm® SA7255P | 12.756 ms | 0 - 79 MB | NPU
|
| 130 |
+
| YOLOv10-Detection | TFLITE | float | Snapdragon® 8 Elite Mobile | 2.042 ms | 0 - 80 MB | NPU
|
| 131 |
+
| YOLOv10-Detection | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.685 ms | 0 - 82 MB | NPU
|
| 132 |
|
| 133 |
## License
|
| 134 |
* The license for the original implementation of YOLOv10-Detection can be found
|