423 5 months ago

A specialized document classification model based on Qwen2.5-VL-3B that automatically detects document types from PDFs and images with high accuracy and calibrated confidence scores.

vision