chamber111/VPPO-7B
Image-Text-to-Text
•
8B
•
Updated
•
42
•
4
VPPO Model
SOTA models for multimodal reasoning, fine-tuned with VPPO. Achieves superior performance by focusing on critical visual tokens.