Hugging Face's logo

qualcomm
/
YOLOv10-Detection

Object Detection
PyTorch
real_time
android
Model card Files
xet
 Community
YOLOv10-Detection
File size: 10,871 Bytes
0c10fbe
 
19f1739
0c10fbe
 
 
3637a56
0c10fbe
 
 
 
 
83508ae
0c10fbe
 
 
83508ae
6103c1f
83508ae
 
 
 
 
6103c1f
83508ae
 
 
 
6103c1f
83508ae
 
 
 
 
 
 
 
ecb0ede
83508ae
 
ecb0ede
 
83508ae
 
 
 
4f145c7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0c10fbe
 
bf7f204
 
0c10fbe
 
 
 
 
 
1f12b49
0c10fbe
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
---
library_name: pytorch
license: agpl-3.0
tags:
- real_time
- android
pipeline_tag: object-detection

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yolov10_det/web-assets/model_demo.png)

# YOLOv10-Detection: Optimized for Qualcomm Devices

Ultralytics YOLOv10 is a machine learning model that predicts bounding boxes and classes of objects in an image.

This is based on the implementation of YOLOv10-Detection found [here](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/detect).
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/v0.62.2/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).

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
Due to licensing restrictions, we cannot distribute pre-exported model assets for this model.
Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.62.2/src/qai_hub_models/models/yolov10_det) 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 [YOLOv10-Detection on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.62.2/src/qai_hub_models/models/yolov10_det) for usage instructions.


## Model Details

**Model Type:** Model_use_case.object_detection

**Model Stats:**
- Input resolution: 640x640
- Model checkpoint: YOLOv10-N
- Model size (float): 8.95 MB
- Model size (w8a16): 3.04 MB
- Model size (w8a8): 2.55 MB
- Number of parameters: 2.33M

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| YOLOv10-Detection | ONNX | float | Snapdragon® X2 Elite | 3.355 ms | 5 - 5 MB | NPU
| YOLOv10-Detection | ONNX | float | Snapdragon® X Elite | 6.674 ms | 5 - 5 MB | NPU
| YOLOv10-Detection | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 3.926 ms | 0 - 242 MB | NPU
| YOLOv10-Detection | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 14.873 ms | 5 - 213 MB | NPU
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 6.678 ms | 5 - 8 MB | NPU
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 6.333 ms | 5 - 9 MB | NPU
| YOLOv10-Detection | ONNX | float | Qualcomm® QCS8450 | 14.873 ms | 5 - 213 MB | NPU
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 7.628 ms | 5 - 7 MB | NPU
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 6.674 ms | 5 - 5 MB | NPU
| YOLOv10-Detection | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 4.385 ms | 1 - 200 MB | NPU
| YOLOv10-Detection | ONNX | float | Snapdragon® 8 Elite Mobile | 4.385 ms | 1 - 200 MB | NPU
| YOLOv10-Detection | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.033 ms | 1 - 220 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® X2 Elite | 2.435 ms | 2 - 2 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® X Elite | 5.626 ms | 2 - 2 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 3.352 ms | 0 - 103 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 6.653 ms | 3 - 102 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 16.993 ms | 2 - 5 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 4.79 ms | 2 - 6 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 5.356 ms | 2 - 6 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® QCS8450 | 6.653 ms | 3 - 102 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 5.813 ms | 2 - 5 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 5.626 ms | 2 - 2 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 25.497 ms | 3 - 196 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 6.646 ms | 3 - 200 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 2.529 ms | 1 - 80 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 2.529 ms | 1 - 80 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 2.284 ms | 0 - 80 MB | NPU
| YOLOv10-Detection | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 6.646 ms | 3 - 200 MB | NPU
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 7.923 ms | 9 - 386 MB | NPU
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Qualcomm® Dragonwing™ IQ-8275 | 11.627 ms | 8 - 12 MB | NPU
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 11.218 ms | 8 - 22 MB | NPU
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Qualcomm® Dragonwing™ IQ-9075 | 12.236 ms | 8 - 11 MB | NPU
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Qualcomm® Dragonwing™ Q-8750 | 6.023 ms | 7 - 357 MB | NPU
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Elite Mobile | 6.023 ms | 7 - 357 MB | NPU
| YOLOv10-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 6.973 ms | 6 - 365 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Snapdragon® X2 Elite | 2.675 ms | 5 - 5 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Snapdragon® X Elite | 4.536 ms | 5 - 5 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 3.034 ms | 0 - 210 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 8.523 ms | 5 - 201 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 5.181 ms | 5 - 12 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.146 ms | 5 - 7 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® SA8775P | 5.592 ms | 2 - 188 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® SA8650P | 5.592 ms | 2 - 188 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® SA8255P | 5.592 ms | 2 - 188 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® QCS8450 | 8.523 ms | 5 - 201 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 5.36 ms | 7 - 13 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 4.536 ms | 5 - 5 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 2.274 ms | 5 - 190 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® SA7255P | 13.297 ms | 0 - 181 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Qualcomm® SA8295P | 8.918 ms | 0 - 169 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 2.274 ms | 5 - 190 MB | NPU
| YOLOv10-Detection | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.989 ms | 5 - 199 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 2.183 ms | 2 - 2 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Snapdragon® X Elite | 4.599 ms | 2 - 2 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 2.919 ms | 2 - 198 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 4.12 ms | 0 - 5 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.236 ms | 2 - 4 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® SA8775P | 4.832 ms | 1 - 176 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® SA8650P | 4.832 ms | 1 - 176 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® SA8255P | 4.832 ms | 1 - 176 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 4.59 ms | 2 - 6 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 4.599 ms | 2 - 2 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 20.693 ms | 2 - 180 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 4.41 ms | 2 - 177 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 2.017 ms | 2 - 178 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Qualcomm® SA7255P | 7.937 ms | 0 - 173 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 2.017 ms | 2 - 178 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 1.747 ms | 2 - 184 MB | NPU
| YOLOv10-Detection | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 4.41 ms | 2 - 177 MB | NPU
| YOLOv10-Detection | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 2.685 ms | 0 - 109 MB | NPU
| YOLOv10-Detection | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 7.404 ms | 0 - 84 MB | NPU
| YOLOv10-Detection | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 4.854 ms | 0 - 14 MB | NPU
| YOLOv10-Detection | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 3.767 ms | 0 - 2 MB | NPU
| YOLOv10-Detection | TFLITE | float | Qualcomm® SA8775P | 5.233 ms | 0 - 82 MB | NPU
| YOLOv10-Detection | TFLITE | float | Qualcomm® SA8650P | 5.233 ms | 0 - 82 MB | NPU
| YOLOv10-Detection | TFLITE | float | Qualcomm® SA8255P | 5.233 ms | 0 - 82 MB | NPU
| YOLOv10-Detection | TFLITE | float | Qualcomm® QCS8450 | 7.404 ms | 0 - 84 MB | NPU
| YOLOv10-Detection | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 4.979 ms | 0 - 13 MB | NPU
| YOLOv10-Detection | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 2.046 ms | 0 - 88 MB | NPU
| YOLOv10-Detection | TFLITE | float | Qualcomm® SA7255P | 12.793 ms | 0 - 77 MB | NPU
| YOLOv10-Detection | TFLITE | float | Qualcomm® SA8295P | 8.252 ms | 0 - 56 MB | NPU
| YOLOv10-Detection | TFLITE | float | Snapdragon® 8 Elite Mobile | 2.046 ms | 0 - 88 MB | NPU
| YOLOv10-Detection | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.687 ms | 0 - 90 MB | NPU

## License
* The license for the original implementation of YOLOv10-Detection can be found
  [here](https://github.com/ultralytics/ultralytics/blob/main/LICENSE).

## References
* [YOLOv10: Real-Time End-to-End Object Detection](https://arxiv.org/abs/2405.14458)
* [Source Model Implementation](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/detect)

## 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).