AI Mini Computer (AMC)-LeapFive

Nickname | quit

Home page

Terminal Devices

AI Mini Computer (AMC)

Product Description

Parameter

Target Applications

Terminal Devices

Applications

Documentation

Features

Content 1

Content 2

Content 3

Overview

The LeapFive AMC (AI Mini Computer) is an AI computing integrated device developed based on the NB2, designed as a deep learning inference tool for various edge applications. With 4 TOPS of computing power, the NB2 supports functionalities such as face detection, recognition, facial expression analysis, object detection, license plate recognition, and voiceprint recognition. The AMC enhances traditional products with deep learning capabilities and is widely applicable to emerging fields such as smart industrial control systems, robotics, industrial computers, smart toys, and intelligent teaching tools.

Features

Features

Core Specifications

• CPU: 4x 64-bit RISC-VCore@1.5 GHz

• Computing Power: 2/4 TOPS

• Memory: 2/4/8GB DDR4/LPDDR4

• Storage: 8/16/32GB eMMC

Support Models

• SSD

• Yolov5

• Deplabv3

• Mobilenetv1v2

• Resnet

• Repvgg

Physical Interface

• USB 3.0（Not compatible with USB 2.0）

Operating Conditions

• Operating Temperature: -10~40℃

• Operating Humidity: 5%~90%, non-condensing

AI teaching aids

Practical teaching tools help integrate theory with practice. A variety of domestically produced autonomous main control boards and terminals can seamlessly adapt to AMC’s neural network-accelerated AI inference capabilities. Choose from a vast array of AI skills, with support for hardware customization and deployment of multi-model, high-precision models.

NB2 (LF566)

LeapFive® LF566 is a high-end edge AI processor built on the RISC-V architecture, specifically designed for edge computing, machine learning, vision, and voice applications.

Next-Generation AI Education

The AI Education Terminal, powered by the NB2, functions as the core controller, enabling the capture and processing of key data in educational scenarios through integrated modules for data annotation, algorithm training, and model management. With support for deploying diverse AI algorithms, the system is tailored for smart educational devices, research training, and teaching assistance platforms. It drives advancements in machine learning and deep learning applications within educational and research contexts.