XNews

Tencent WeChat open-sources WeVisDoc, an end-to-end document parser

Tencent WeChat open-sources WeVisDoc, an end-to-end document parser

Tencent WeChat's vision team released WeVisDoc, an end-to-end document parser that converts page images directly to structured Markdown. Based on Qwen3-VL, it offers 4B and 2B versions, supports Chinese and English, and uses a two-stage training strategy with residual analysis to improve performance on degraded documents.

Source posts