nndeploy is an end-to-end model deployment framework. Based on multi-terminal inference and directed acyclic graph model deployment, it is committed to providing users with a cross-platform, easy-to-use, and high-performance model deployment experience.

an edge-real-time anchor-free object detector with decent performance

YOLOv5 in PyTorch > ONNX > RKNN

Run your yolov7 object detection with Rockchip NPU platforms (RK3566, RK3568, RK3588, RK3588S, RV1103, RV1106, RK3562).

Inference YOLOv8 segmentation on ONNX, RKNN, Horizon and TensorRT

simple yolov5 rtspserver for rk3588

A demo show how to inference blazeface model on Rockchip RV1126.

RKNN模型推理部署模板

基于u2net网络进行简单修改使其部署到rk3588板子上

Inference YOLOv8 detection on ONNX, RKNN, Horizon and TensorRT

Deploy deep learning model on difference hardware and framework. (TensorRT/ONNX/MNN/RKNN)

export any your YOLOv7 model to TensorFlow, TensorFlowJs, ONNX, OpenVINO, RKNN,...

Automated script to convert Huggingface and GGUF models to rkllm format for running on Rockchip NPU

Superresolution running on Rockchip NPU (RK3588, etc..)

Web chat front end for rk3588_npu_llm_server / RK3588 LLM chat interface

fatigue detect rknn/onnx model deploy in rk3568 npu（ROCK 3A）

瑞芯微芯片的rknn推理框架部署(yolo模型)