State-of-the-art on-device vision model specializing in complex document parsing and high-resolution spatial reasoning.

Outputs 2

MiniCPM-V 4.5

model
Architecture DENSE

MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipe

paper

arXiv: 2509.18154

multimodalon-devicevision

Related