SpatioLM is a parameter-efficient framework for improving spatial intelligence in vision-language models (VLMs). It introduces a plug-and-play spatio-
-
xiaomi-research/SpatioLM-Understanding-InternVL3.5
Image-Text-to-Text • 9B • Updated -
xiaomi-research/SpatioLM-Understanding-SenseNovaSI
Image-Text-to-Text • 8B • Updated -
xiaomi-research/SpatioLM-Perception-InternVL3.5
Image-Text-to-Text • 9B • Updated -
xiaomi-research/SpatioLM-Perception-SenseNovaSI
Image-Text-to-Text • 8B • Updated