Skip to main content

Tencent Merges AI Model Teams, Yao Shunyu Takes the Helm

Tencent is shaking up its AI research structure. On July 23, the company announced it has merged its multimodal model department with its large language model department, forming a new Basic Model Department. The combined unit will be overseen by Yao Shunyu, Tencent's chief AI scientist.

This isn't a sudden decision. Back in December, Yao had already taken charge of the large language model team. Now, by bringing both the multimodal and language lines under one roof, Tencent is signaling a clear intent: to unify the development of models that can both "see" images and "converse" naturally. The goal is to break down the walls between these two areas and create a more cohesive approach to full-modal AI.

The restructuring is designed to improve efficiency and collaboration. Instead of separate teams working on different aspects of AI, the new department will have a single command center. This should speed up development and help Tencent explore the upper limits of what full-modal models can achieve.

For context, multimodal models handle different types of data—like images, text, and audio—while large language models focus on understanding and generating human language. Combining them is a logical step toward more versatile AI systems that can interact with the world in richer ways.

Yao Shunyu, a well-known figure in the AI community, will lead this charge. His background in both research and management makes him a natural fit for the role. Under his guidance, the Basic Model Department will work on advancing Tencent's Yuanmeng technology roadmap, which aims to integrate visual and conversational AI seamlessly.

Industry observers see this as a smart move. By aligning resources and leadership, Tencent can avoid duplication of effort and foster more cross-pollination between teams. It also positions the company to compete more effectively in the fast-evolving AI landscape.

Key Points

  • Tencent merges multimodal and large language model departments into a new Basic Model Department.
  • Yao Shunyu, chief AI scientist, will lead the combined unit.
  • The restructuring aims to improve collaboration and push full-modal AI capabilities.
  • The move builds on Yao's previous role as head of the large language model department.
  • It reflects a broader industry trend toward integrating different AI modalities.