YOLO12-n is the ultra-lightweight (nano) version of the 12th-generation YOLO object detection models, specifically designed for extremely resource-constrained edge devices such as microcontrollers (MCUs), embedded vision chips, and low-power AI terminals. It preserves the core advantages of the YOLO family while adopting structural pruning, low-bit quantization, and lightweight module design to minimize parameter count and computational load.

Despite its compact size, YOLO12-n maintains decent detection accuracy and performs well in low- to mid-complexity object detection tasks. Its ultra-fast inference speed and minimal power consumption make it ideal for smart hardware, wearables, smart home systems, and industrial edge gateways.

YOLO12-n is compatible with major deployment frameworks including TensorRT, TFLite, and ONNX Runtime, and supports INT8/FP16 quantization. It offers an excellent balance of speed, accuracy, and deployment flexibility, making it a strong candidate for real-world edge AI applications.

Source model

• Input shape: 1x3x640x640
• Number of parameters: 2.51M
• Model size: 10.18M
• Output shape: 1x84x8400

Source model repository: yolo12n

When the user has fine-tuned the source model, the model conversion process must be performed again.

Users can refer to either of the following two methods to complete the model conversion:

• Using AIMO for model conversion: Click Model Conversion Reference in the Performance Reference section on the right to view the conversion steps.

• Using Qualcomm QNN for model conversion: Please refer to the Qualcomm QNN Documentation.

The model performance benchmarks and example code provided by Model Farm are all implemented based on the APLUX AidLite SDK.

For models in .bin format, you can use either of the following two inference engines to run inference on Qualcomm chips:

Inference using APLUX AidLite: please refer to the APLUX AidLite Developer Documentation

Inference using Qualcomm QNN: Please refer to the Qualcomm QNN Documentation

Inference Example Code

The inference example code is implemented using the AidLite SDK.

Click Model & Code to download the model files and the inference code package. The file structure is as follows:

/model_farm_{model_name}_aidlite
    
    |__ models # folder where model files are stored

    |__ python # aidlite python model inference example

    |__ cpp # aidlite cpp model inference example

    |__ README.md