NPU Usage
📄️ Quick Start
📄️ RKNN Installation
Get started with RKNN installation and experience efficient AI model inference on Rockchip NPU
📄️ RKNN Model Zoo
📄️ RKLLM Installation
📄️ RKLLM Usage
📄️ RKNN Quick Start
📄️ RKNN Toolkit2 PC
Use the RKNN toolkit to explore simulated inference of AI models and experience the efficiency and precision of intelligent image segmentation
📄️ RKNN Toolkit Lite2 YOLOv5
📄️ RKNN Ultralytics YOLOv11
Deploy YOLOv11 on RK3588/RK356X boards, ushering in a new era of intelligent object detection where technology and human care merge in precise recognition
📄️ RKNN Custom YOLO
Convert self-trained YOLO models
📄️ YOLOv8
📄️ PP-YOLOE
📄️ YOLO World
📄️ RetinaFace
📄️ YOLOv8-Seg
📄️ DeepLabV3
📄️ MobileSAM
📄️ MobileNet
📄️ ResNet
📄️ PP-OCR
📄️ LPRNet
📄️ CLIP
📄️ Whisper
📄️ Wav2Vec 2.0
📄️ Kiln: local LLM + vision on the NPU (mainline kernel)
Use Kiln on the ROCK 4D to run the vendor RKLLM/RKNN NPU stack on a mainline Linux kernel: one command for a local LLM chat, image recognition, and an OpenAI-compatible API you can point Open WebUI at.