Qwen’s Core Team Shaken as Technical Lead, Researchers Exit

Several key figures behind Alibaba’s Qwen large language model project are exiting the company, raising questions about leadership changes within one of China’s most prominent AI initiatives.

Among the most notable exits is Junyang Lin, the technical lead of the Qwen project and one of the central architects behind Alibaba’s push into open-source generative AI. Lin announced his departure in a brief post on X, writing that he was “stepping down” and bidding farewell to the project he helped build. 

The announcement came shortly after the launch of Alibaba’s latest Qwen 3.5 small model series, a set of lightweight multimodal models designed for on-device AI and agent applications.

Chen Cheng, a contributor for Qwen models, wrote on X, “I'm truly heartbroken. I know leaving wasn't your choice. Just last night, we were side by side launching the Qwen3.5 small model.” 

“I honestly can't imagine Qwen without you,” they added. 

Other researchers associated with the Qwen project have also recently announced their departures. 

Kaixin Li and Binyuan Hui have also departed from the company. Developers and users on social media expressed surprise at the news following the announcements.

“End of an era. Qwen lost its tech lead,” Yuchen Jin, Co-founder and CTO of decentralised AI computing and cloud platform Hyperbolic Labs, wrote in a post on X. “Thank you for pushing open source AI forward.”

Nathan Lambert, a researcher at Allen AI, wrote on X, “The gaping hole that [Qwen's] imploding would leave in the open research ecosystem will be hard to fill. The small models are irreplaceable.”

There has been no official statement from Qwen regarding these departures, and the reasons behind the exits remain unclear, leading to speculation on social media. There are also unconfirmed rumours circulating that executives who previously worked on Google’s Gemini may step in to fill these positions.