LG unveiled exaeone Deep artificial intelligence; ChatGpt more powerful?

LG has recently unveiled its new model of artificial intelligence, the Exaeone Deep, which can in some respects abandon popular and powerful models such as GPT, Jina and Deepsic.

What is the capabilities of the LG Exaone Deep Artificial Intelligence?

Building automatic artificial intelligence tools that can make decisions without the need for human supervision requires the acquisition of powerful models of reasoning, and large companies such as Openai have understood the importance of this issue to develop rational models. A few companies are capable of engineering such models, and now LG has joined them with the introduction of Exaone Deep artificial intelligence. The model, which is the product of the efforts of the Korean Technology Giant’s LG AI Research section, can showcase the ability to argue and have high potential for industrial and commercial use.

In the preview of the Exaone Deep Artificial Intelligence, LG is capable of understanding mathematical logic, reasoning about scientific concepts, solving programming problems, and advanced reasoning, and shows that this tool can be used in various branches of science and mathematics alongside coding. After the official launch of the new LG model, we see it on the list of Notable AI Models, which confirms its high power. It is noteworthy that the Exaone Deep 32B is the first Korean model in the last two years to be among the top models of artificial intelligence.

LG’s new artificial intelligence has achieved the highest score in the CSAT and AIME math tests and topped the list with a slight difference than the Deepseek-R1 (671b). It has also been able to perform better with a tangible difference in all mathematical tests compared to the O1-Mini model. It is noteworthy that the CSAT test is held every year to select the US Mathematics Olympiad and raises conceptual mathematics issues that require a deep rational understanding and high ability to reason.

The GPQA Diamond test is another criterion for measuring the power of artificial intelligence -based tools that evaluates problem -solving performance at a PHD level in the field of physics, chemistry and biology. The LiveCodebench test is also used to investigate the power of coding models. The Exaone Deep AI has been able to get out of both tests and perform better than the O1-Mini and QWQ-32B models. However, at the highest parameter level, the Dip-Six R1-671B model has been able to perform better in coding and answering various science questions.

You can apply for the new LG artificial intelligence test via the trial version on the Hugging Face platform.