Python
Skywork-R1V3 - When Images and Text Work Together
Skywork-R1V3 is a 38B multimodal model combining visual perception with deep reasoning. It leads benchmarks at 76% MMMU and 77.1% MathVista accuracy.
3,160
Language
HomeLanguages
Sections
Skywork-R1V3 is a 38B multimodal model combining visual perception with deep reasoning. It leads benchmarks at 76% MMMU and 77.1% MathVista accuracy.