Samsung Electronics recently hosted the Samsung Developer Conference Korea 2024, marking its 11th edition aimed at engaging software developers.

At the conference, Samsung unveiled its second-generation AI model, Samsung Gauss2, which promises enhanced performance and new applications across various data types.

The model is multimodal, enabling it to process images, text, and code simultaneously, thereby improving functionality across different devices.

The Supreme model employs a 'Mixture of Experts' technology, activating the most suitable expert model based on task requirements to deliver high performance.

Gauss2 supports up to 14 languages and various programming languages, further enhancing its versatility and application in diverse settings.

The processing speed of Gauss2 leads to reduced costs and shorter service wait times, making it a competitive option in the AI landscape.

Despite these advancements, challenges remain for Samsung, particularly competition from China in sectors like mobile phones and memory chips.

Gauss2 is available in three distinct sizes: Compact, Balanced, and Supreme, each tailored for specific use cases.

The Balanced and Supreme models significantly outperform many open-source AI models, achieving processing speeds 1.5 to 3 times faster in language generation and coding tasks.

Additionally, Gauss2 is utilized in Samsung's call centers to categorize and summarize customer interactions, enhancing efficiency and service quality.

Paul Kyungwhoon Cheun, President and CTO of Samsung's Device eXperience Division, emphasized the company's commitment to integrating advanced AI to enhance user experiences.

Internally, Gauss2 is already in use at Samsung, with over 60% of developers in the Device eXperience division utilizing it for coding, translating, and summarizing tasks.