We introduce Yuan3.0 Flash, an open-source Mixture-of-Experts (MoE) MultiModal Large Language Model featuring 3.7B activated parameters and 40B total parameters, specifically designed to enhance performance on enterprise-oriented tasks while maintaining competitive capabilities on general-purpose tasks. To address the overthinking phenomenon commonly observed in Large Reasoning Models (LRMs), we propose Reflection-aware Adaptive Policy Optimization (RAPO), a novel RL training algorithm that effectively regulates overthinking behaviors. In enterprise-oriented tasks such as retrieval-augmented generation (RAG), complex table understanding, and summarization, Yuan3.0 Flash consistently achieves superior performance. Moreover, it also demonstrates strong reasoning capabilities in domains such as mathematics, science, etc., attaining accuracy comparable to frontier model while requiring only approximately 1/4 to 1/2 of the average tokens. Yuan3.0 Flash has been fully open-sourced to facilitate further research and real-world deployment: https://github.com/Yuan-lab-LLM/Yuan3.0.
@article{arxiv.2601.01718,
title = {Yuan3.0 Flash: An Open Multimodal Large Language Model for Enterprise Applications},
author = {YuanLab. ai and : and Shawn Wu and Sean Wang and Louie Li and Darcy Chen and Allen Wang and Jiangang Luo and Xudong Zhao and Joseph Shen and Gawain Ma and Jasper Jia and Marcus Mao and Claire Wang and Hunter He and Carol Wang and Zera Zhang and Jason Wang and Chonly Shen and Leo Zhang and Logan Chen and Qasim Meng and James Gong and Danied Zhao and Penn Zheng and Owen Zhu and Tong Yu},
journal= {arXiv preprint arXiv:2601.01718},
year = {2026}
}