Former OpenAI Researcher Tian Yonglong to Join Tencent Hunyuan Multimodal Team
A significant personnel move has surfaced in the AI industry: Yonglong Tian, previously a researcher at OpenAI, has joined Tencent. He is expected to lead the multimodal model direction at Tencent Hunyuan, focusing on vision-language models (VLMs). While not yet officially confirmed, the news has generated considerable attention. If finalized, Tian would become the latest core research talent to join the Tencent Hunyuan team, following Shunyu Yao.
Tian's arrival underscores Tencent's ongoing commitment to multimodal large models. He and Tencent's current Chief AI Scientist, Shunyu Yao, previously worked together at OpenAI, witnessing the evolution of GPT and multimodal technologies. With a strong background in computer vision, Tian earned his bachelor's degree from Tsinghua University, a master's from the MMLab at the Chinese University of Hong Kong, and a PhD from MIT. His research trajectory shows a clear shift from traditional visual recognition to generative models and representation learning, earning nearly 30,000 citations on Google Scholar.

Academically, Tian has made notable contributions. His proposal of "Supervised Contrastive Learning" extended contrastive learning to supervised settings, significantly influencing visual pre-training model design. More recently, his work on autoregressive image generation—especially the Fluid series—demonstrated the advantages of continuous tokens for visual quality, offering new pathways for large model generation. Prior to Tencent, he worked at Google Research and Google DeepMind before joining OpenAI as a technical staff member.
This hire is a key element of the recent restructuring and talent strategy for Tencent Hunyuan's multimodal team. Since Tencent officially formed the Large Language Model Department and Multimodal Model Department in April 2025, the team has been actively expanding. Several experienced researchers from Microsoft, Alibaba, and Tsinghua University have joined, aiming to resolve past issues of scattered resources and build a more competitive multimodal R&D system.
With the addition of top-tier talent like Tian, Tencent Hunyuan's research capabilities in multimodal and visual perception are poised to strengthen further. As a major player in the AI industry, Tencent's ongoing talent acquisitions signal its deep strategic commitment to advancing the multimodal capabilities of the Hunyuan model.
Related article
Claude Adds Multilingual Voice Support, Bringing Real Phone Calls Closer
The voice interaction experience of artificial intelligence is about to undergo a major update. According to the latest news from tech media Android Authority, Anthropic is preparing a significant upgrade for its AI assistant Claude, focusing on brea
Apple Store App Unveils AI Virtual Shopping Assistant to Support Product Q&A and Decision-Making
MacRumors reports that Apple Inc. has recently revised the privacy policy for its Apple Store app, signaling the imminent arrival of a "virtual shopping assistant." Designed to answer product inquiries and guide purchasing decisions via a generative
Apple Watch’s New AI Features Normalize Always-Listening Tech
Apple is pushing consumers to embrace a new norm: your devices are constantly listening.During Wednesday’s “Surprise and Shine” iPhone launch, which unveiled the latest hardware, the true revelation wasn’t the rumored foldable iPhone, but rather how
Related Special Topic Recommendations
Comments (0)
0/500
A significant personnel move has surfaced in the AI industry: Yonglong Tian, previously a researcher at OpenAI, has joined Tencent. He is expected to lead the multimodal model direction at Tencent Hunyuan, focusing on vision-language models (VLMs). While not yet officially confirmed, the news has generated considerable attention. If finalized, Tian would become the latest core research talent to join the Tencent Hunyuan team, following Shunyu Yao.
Tian's arrival underscores Tencent's ongoing commitment to multimodal large models. He and Tencent's current Chief AI Scientist, Shunyu Yao, previously worked together at OpenAI, witnessing the evolution of GPT and multimodal technologies. With a strong background in computer vision, Tian earned his bachelor's degree from Tsinghua University, a master's from the MMLab at the Chinese University of Hong Kong, and a PhD from MIT. His research trajectory shows a clear shift from traditional visual recognition to generative models and representation learning, earning nearly 30,000 citations on Google Scholar.

Academically, Tian has made notable contributions. His proposal of "Supervised Contrastive Learning" extended contrastive learning to supervised settings, significantly influencing visual pre-training model design. More recently, his work on autoregressive image generation—especially the Fluid series—demonstrated the advantages of continuous tokens for visual quality, offering new pathways for large model generation. Prior to Tencent, he worked at Google Research and Google DeepMind before joining OpenAI as a technical staff member.
This hire is a key element of the recent restructuring and talent strategy for Tencent Hunyuan's multimodal team. Since Tencent officially formed the Large Language Model Department and Multimodal Model Department in April 2025, the team has been actively expanding. Several experienced researchers from Microsoft, Alibaba, and Tsinghua University have joined, aiming to resolve past issues of scattered resources and build a more competitive multimodal R&D system.
With the addition of top-tier talent like Tian, Tencent Hunyuan's research capabilities in multimodal and visual perception are poised to strengthen further. As a major player in the AI industry, Tencent's ongoing talent acquisitions signal its deep strategic commitment to advancing the multimodal capabilities of the Hunyuan model.
Claude Adds Multilingual Voice Support, Bringing Real Phone Calls Closer
The voice interaction experience of artificial intelligence is about to undergo a major update. According to the latest news from tech media Android Authority, Anthropic is preparing a significant upgrade for its AI assistant Claude, focusing on brea
Apple Store App Unveils AI Virtual Shopping Assistant to Support Product Q&A and Decision-Making
MacRumors reports that Apple Inc. has recently revised the privacy policy for its Apple Store app, signaling the imminent arrival of a "virtual shopping assistant." Designed to answer product inquiries and guide purchasing decisions via a generative
Apple Watch’s New AI Features Normalize Always-Listening Tech
Apple is pushing consumers to embrace a new norm: your devices are constantly listening.During Wednesday’s “Surprise and Shine” iPhone launch, which unveiled the latest hardware, the true revelation wasn’t the rumored foldable iPhone, but rather how





Home






