This collection consists of SpatialThinker 3B and 7B model checkpoints, and STVQA-7K, a Spatial VQA dataset used for training the models.
-
SpatialThinker: Reinforcing 3D Reasoning in Multimodal LLMs via Spatial Rewards
Paper • 2511.07403 • Published • 14 -
OX-PIXL/STVQA-7K
Viewer • Updated • 7.59k • 39 • 2 -
OX-PIXL/SpatialThinker-3B
Image-Text-to-Text • 4B • Updated • 74 • 3 -
OX-PIXL/SpatialThinker-7B
Image-Text-to-Text • 8B • Updated • 31 • 2