On the morning of December 17, at the 2025 Xiaomi Human-Car-Home Ecosystem Partners Conference, Lu Weibing, Partner and President of Xiaomi Group, delivered a keynote speech titled "Journey Together, Surging Future."
He announced that Xiaomi's self-developed AI model, Xiaomi MiMo-V2-Flash, has been officially open-sourced. Lu described it as a new language foundation for the Agent era.
According to the timeline of Xiaomi's self-developed MiMo series shared by Lu, the company has already released several models, including the reasoning model MiMo-7B, the visual reasoning model MiMo-VL, the native end-to-end audio generation model MiMo-Audio, the on-device visual-language model MiMo-VL-Miloco, and the embodied model MiMo-Embodied.
Comments