K230

Sign inSign up

K230

K230

The K230 chip is an end-side AIoT (Artificial Intelligence of Things) chip based on the RISC-V architecture. It features high precision, low latency, high performance, ultra-low power consumption, and fast boot-up, making it widely applicable to various intelligent product scenarios such as edge-side large model multimodal access terminals, 3D structured light depth perception modules, interactive robots, open-source hardware, smart manufacturing, smart homes, smart education hardware, and many other fields.

Function Diagram

Function Diagram

Device Parameter

CPU
- CPU 1: RISC-V processor, 1.6GHz, 32KB I-cache, 32KB D-cache, 256KB L2 Cache, 128-bit RVV 1.0 extension
- CPU 0: RISC-V processor, 0.8GHz, 32KB I-cache, 32KB D-cache, 128KB L2 Cache
KPU
- Supports INT8 and INT16
- Typical network performance: Resnet 50 ≥ 85fps @INT8; Mobilenet_v2 ≥ 670fps @INT8; YoloV5S ≥ 38fps @INT8
DPU
- 3D structured light depth engine, supporting a maximum resolution of 1920*1080
VPU
- H.264 and H.265 encoder and decoder with a maximum resolution support of 4096*4096
- Encoder performance: 3840*2160@20fps
- Decoder performance: 3840*2160@40fps
- JPEG codec: supports up to 8K (8192*8192) resolution
Image Input
- Supports up to three lanes MIPI CSI (1x4lane + 1x2lane) or 3x2lane
Display Output
- 1 lane MIPI DSI, 1x4 lane or 1x2 lane
- Maximum resolution 1920*1080@60fps
On-chip Interfaces
- 5 x UART
- 5 x I2C
- 6 x PWM
- 64 x GPIO + 8 x PMU GPIO
- 2 x USB 2.0 OTG
- 2 x SDxC: SD3.01, eMMC 5.0
- 3 x SPI: 1 x OSPI + 2 x QSPI
- WDT / RTC / Timer

Application Scenario

Assisted Driving

Assisted Driving

Smart Network Camera

Smart Network Camera

Robot Interaction

Robot Interaction

Unmanned Aerial Vehicle

Unmanned Aerial Vehicle

Face access control

Face access control

Industrial Camera

Industrial Camera

Smart Power Equipment

Smart Power Equipment

STEAM Education

STEAM Education

SDK List

---Choose the best fit from four software options and click to view the guide.---

Simple and intuitive, easy to learn, quick to start

Performance is slightly lower compared to others

Real-time RTOS, with rapid boot speed

Has a certain learning curve

Multifunctional, with a rich ecosystem

Not as quick to get started for beginners as CanMV

Linux+RT-Smart SDK

Rich ecosystem

Challenging for newcomers

Document

By integrating Micropython on the K230, you can conveniently and quickly develop AI applications using the Micropython language.

Based on the Linux system, developed using C/C++ languages, this includes tutorials, SDK usage instructions, application example explanations, and other materials.

Based on the RT-Smart system, developed using C/C++ languages, this includes tutorials, SDK usage instructions, application example development guides, API references, and other related materials.

Linux+RT-Smart SDK

Developed on a dual-core heterogeneous system of Linux + RT-smart, using C/C++ languages. This includes quick start guides, SDK usage instructions, application example development tutorials, API references, and other relevant resources.

AI Development Tutorials

This document introduces instance tutorials for K230 AI application development, including the use of companion tools and other related resources. There are multiple development methods available, such as using an online cloud training platform, AI Cube, or KTS for source code-level AI applications.

Hardware Design Materials

Includes hardware design guide documents LP3, LP4, CanMV engineering design files

Download

Includes images related to Linux+RTOS, Linux, RTOS, CanMV

Includes chip pin definitions, OTP configuration tools, datasheets, hardware design guides, and hardware documentation usage instructions

Includes Toolchain, CanMV IDE, download tools, Debug Tools, etc.