Ant and Renmin University of China release the first native MoE diffusion language model.

At the 2025 InclusionThe Bund Conference, Ant Group and Renmin University of China jointly released the industry's first native MoE architecture diffusion language model "LLaDA-MoE". LLaDA-MoE completed the diffusion language model with MoE architecture training from scratch on about 20T data, verifying the scalability and stability of industrial-scale training. The model will be fully open-sourced in the near future to promote the technological development of the global AI community in dLLM.