ThunderSoft brings two new large-scale products to IOTE 2023

Today, the "E 2023 20th International Exhibition" opened grandly. Thundersoft brought its Rubik's Cube large models, solutions, industry solutions and other achievements to the exhibition, and officially launched two new large models, the Rubik's Cube Brain and the Rubik's Cube Kit, which received widespread attention in the industry.

Nowadays, applications based on big models have become a new engine for improving enterprise efficiency and an important tool for promoting the process of intelligence.

Thundersoft Rubik's Cube Brain is an enterprise smart office software and hardware integrated platform built based on Thundersoft Rubik's Cube big model. The platform can not only realize basic capabilities such as natural language interaction, logical reasoning, text generation and knowledge, but also continuously evolve and support third-party big model applications, enabling enterprises to quickly access GC and quickly adapt to enterprise customization needs.

In order to meet the diverse requirements of enterprises of different sizes, ThunderSoft has launched three Rubik's Cube Brain products with different computing power and levels: Mars, Venus and Jupiter. Through the pre-installed multiple enterprise applications and full utilization of resources, the Rubik's Cube Brain can be "ready to use out of the box", saving enterprises time and financial investment in model selection, private environment construction, and application construction, allowing enterprises to quickly build exclusive large model capabilities while saving costs.

According to the test results, with the empowerment of Magic Cube Brain, the efficiency of knowledge acquisition of enterprises can be increased by more than 6 times, the efficiency of office processes can be increased by more than 5 times on average, and 30% of communication costs can be saved.

In addition, the application of Magic Cube Brain can also lower the professional threshold of creation and similar tools, thereby increasing the creation efficiency by more than 10 times, providing a strong technical foundation for the smart office of enterprises.

The Magic Cube Brain Platform can quickly build enterprise-specific large model capabilities

Another large model product launched this time, Chuangda Rubik's Cube Suite, is an AI tool developed based on the Rubik's Cube large model. It can realize functions such as code completion, code generation, code monitoring and solution generation. It can provide customers with a full-stack AI assistant service from code development requirements to solution delivery, and can provide expert advice and guidance during the code development process, thereby helping software developers to comprehensively improve development efficiency. The Rubik's Cube Suite now supports multiple mainstream programming languages ​​such as , c/, and .

In addition, with the continuous accumulation of code knowledge base and the addition of industry codes such as the Internet of Things, the accuracy of the generated code of the Magic Cube Suite will continue to improve.

The localized deployment method also allows users to enjoy faster code generation response speed while ensuring code security. Tests show that the application of the Magic Cube Suite can improve users' work efficiency by an average of 20%.

Chuangda Rubik's Cube Kit——AI programming tool developed based on the Rubik's Cube large model

During the exhibition, on display simultaneously with the Rubik's Cube Brain and Rubik's Cube Kit were the Rubik's Cube Large Language Model (Edge Edition), Rubik's Cube Sales Assistant, Rubik's Cube Speaker, Rubik's Cube Butler, and other large models of the Rubik's Cube. Their applications and practical results in different scenarios such as the edge side, terminal side, and home, attracted widespread attention from the audience.

In the future, ThunderSoft will continue to delve deeper into the field of big models, give full play to its profound accumulation in the fields of smart hardware and smart industries, and continue to explore innovative applications of big models in the fields of robots, smart cars, XR, etc., to contribute to the intelligent transformation of thousands of industries.

Review editor: Liu Qing