Khadas Docs products:sbc:edge2:npu https://docs.khadas.com/ 2026-04-18T07:32:52+00:00 Khadas Docs https://docs.khadas.com/ https://docs.khadas.com/ttps://docs.khadas.com/lib/tpl/dokuwiki-new/images/favicon.ico text/html 2025-01-02T04:45:55+00:00 Anonymous (anonymous@undisclosed.example.com) Convert Your Model to ONNX https://docs.khadas.com/products/sbc/edge2/npu/convert-onnx?rev=1735811155&do=diff npu model onnx pytorch tensorflow paddle caffe darknet Convert Your Model to ONNX Why use ONNX There are many deep learning frameworks available. But these frameworks cannot be converted to each other. ONNX acts as a bridge between them. One framework can convert its model to ONNX and then convert ONNX model to the other framework model. text/html 2025-02-13T20:50:34+00:00 Anonymous (anonymous@undisclosed.example.com) Large Model - DeepSeek-R1-Distill-Qwen-1.5B/7B https://docs.khadas.com/products/sbc/edge2/npu/deepseek-r1-distill-qwen-1.5b-7b?rev=1739497834&do=diff deepseek npu edge2 rk3588 Large Model - DeepSeek-R1-Distill-Qwen-1.5B/7B Convert Model Convert model should be done on Linux PC. Convert DeepSeek-R1-Distill-Qwen-1.5B need GPU memory or CPU memory at least 13G. Convert DeepSeek-R1-Distill-Qwen-7B at least 32G. Build virtual environment text/html 2024-04-25T03:46:35+00:00 Anonymous (anonymous@undisclosed.example.com) NPU Demos https://docs.khadas.com/products/sbc/edge2/npu/demos?rev=1714031195&do=diff NPU Demos demos index text/html 2025-04-23T22:22:48+00:00 Anonymous (anonymous@undisclosed.example.com) LLM on Edge2 https://docs.khadas.com/products/sbc/edge2/npu/llm-on-edge2?rev=1745461368&do=diff LLM on Edge2 To quickly run large language models (LLMs) on your Khadas Edge2 (RK3588S) device, simply run the following script: khadas_llm.sh This script will: * Install necessary dependencies such as cmake * Prompt you to choose from the following supported models: text/html 2024-04-25T03:46:48+00:00 Anonymous (anonymous@undisclosed.example.com) NPU Applications https://docs.khadas.com/products/sbc/edge2/npu/npu-applications?rev=1714031208&do=diff NPU Applications Get Source Code Clone the source code form our khadas/edge2-npu. $ git clone https://github.com/khadas/edge2-npu There are two types of application source code in C++ and Python. Python Applications Enter Python directory, $ cd Python Install dependences, text/html 2024-06-06T21:57:57+00:00 Anonymous (anonymous@undisclosed.example.com) Edge2 NPU Model Convert https://docs.khadas.com/products/sbc/edge2/npu/npu-convert?rev=1717725477&do=diff Edge2 NPU Model Convert Build Virtual Environment The SDK only supports python3.6 or python3.8, here is an example of creating a virtual environment for python3.8. Install python packages. $ sudo apt update $ sudo apt install python3-dev python3-numpy text/html 2025-04-23T06:42:38+00:00 Anonymous (anonymous@undisclosed.example.com) Object Detection with RTSP Streaming https://docs.khadas.com/products/sbc/edge2/npu/object-detection-demo-with-rtsp?rev=1745404958&do=diff Object Detection with RTSP Streaming Overview This document describes how to compile and run the object detection demo with RTSP streaming on the Khadas Edge2 using its NPU (Neural Processing Unit). Prerequisites Before building the project, ensure your system is up to date and the required dependencies are installed: text/html 2023-10-23T00:56:25+00:00 Anonymous (anonymous@undisclosed.example.com) Edge2 NPU Notes https://docs.khadas.com/products/sbc/edge2/npu/start?rev=1698036985&do=diff edge2 npu Edge2 NPU Notes The Edge2 SBC has a 6 TOPS npu index About NPU A neural processing unit (NPU) is a microprocessor that specializes in the acceleration of machine learning algorithms, typically by operating on predictive models such as artificial neural networks (ANNs) or random forests (RFs). It is, also, known as a neural processor.