YOLOv10b: Object Detection
YOLOv10b is the large-scale model in the YOLOv10 family, designed for high-precision object detection tasks. Compared to the lightweight and medium variants, YOLOv10b features a deeper network architecture and more parameters, enabling it to capture richer feature representations and significantly improve detection of small objects and complex scenes. The model employs an advanced anchor-free mechanism, combined with multi-scale feature fusion and a powerful decoupled head design, enhancing detection accuracy and robustness. YOLOv10b is suitable for deployment on high-performance servers or advanced edge devices, widely used in autonomous driving, intelligent security, and industrial inspection applications with demanding requirements.
Source model
- Input shape: 1x3x640x640
- Number of parameters: 19.62M
- Model size: 72.99M
- Output shape: 1x300x6
The source model can be found here
Performance Reference
Please search model by model name in Model Farm
Inference & Model Conversion
Please search model by model name in Model Farm