Xiaomi-Robotics-1
XiaomiXiaomi-Robotics-1 adopts a Mixture-of-Transformers (MoT) architecture consisting of a pre-trained vision-language model (VLM) (i.e., Qwen3-VL [3]) and a diffusion transformer (DiT) [57].
1 more from this model
Xiaomi-Robotics-0 adopts a mixture-of-transformers (MoT) model architecture.