license: agpl-3.0
pipeline_tag: object-detection
tags:
- AIoT
- QNN
YOLOv10x: Object Detection
YOLOv10x is the largest and most powerful model in the YOLOv10 series, tailored for scenarios with extremely high demands on detection accuracy. It features a very deep architecture and incorporates state-of-the-art techniques such as an anchor-free detection mechanism, decoupled head design, and multi-scale feature fusion. YOLOv10x achieves exceptional recognition precision and robustness in complex environments while maintaining efficient end-to-end inference. It is ideal for high-stakes applications like autonomous driving, smart cities, and precision industrial vision where top-tier accuracy and reliability are critical.
Source model
- Input shape: 1x3x640x640
- Number of parameters: 30.34M
- Model size: 112.73M
- 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