TENCENT recently announced a comprehensive restructuring of its Hunyuan large model R&D unit, further accelerating its strategic deployment in the AI field. As part of this adjustment, Tencent has established two new departments — the Large Language Model Department and the Multimodal Model Department. In addition to the creation of these new departments, Tencent has also strengthened its large model data capabilities and platform infrastructure. The Data Platform Department will focus on end-to-end management and development of large model data, while the Machine Learning Platform Department will concentrate on building a unified platform integrating machine learning and big data, providing a comprehensive and efficient PaaS (Platform as a Service) foundation for AI model training, inference, and big data operations. Prior to this, Tencent had already integrated four major AI product lines, including Yuanbao, IMA, QQ Browser, and Sogou Input Method, advancing a multi-model strategy of “core technology self-development plus embracing advanced open-source solutions.” Since the beginning of this year, Tencent has significantly accelerated the development of its self-developed Hunyuan large model, launching the fast-thinking model TurboS and the deep-thinking model T1. Meanwhile, Tencent has also released multiple new versions of models in video generation and 3D generation, such as Hunyuan 3D Generation, Video Generation, DiT Text-to-Image, and a trillion-parameter MoE (Mixture of Experts) language model, all of which have been open-sourced. To better support this strategic layout, Tencent is also ramping up its recruitment efforts in the AI field. Last month, Tencent announced its largest-ever employment plan, aiming to add 28,000 internship positions over the next three years while increasing conversion rates for full-time hires. By 2025, Tencent expects to welcome 10,000 campus-recruited interns.(SD News) | 