Tag
WeVisDoc is an end-to-end document parser fine-tuned from Qwen3-VL models that converts page images to structured Markdown with LaTeX and HTML, achieving top performance on benchmarks like OmniDocBench v1.6.