This dataset contains OCR results from images in davanstrien/encyclopaedia-britannica-1771 using GLM-OCR, a compact 0.9B OCR model achieving SOTA performance.

Processing Details

Source Dataset: davanstrien/encyclopaedia-britannica-1771 Model: zai-org/GLM-OCR Task: text recognition Number of Samples: 2,724 Processing Time: 376.7 min Processing Date: 2026-02-18 20:29 UTC

Configuration

Image Column: image Output Column: markdown… See the full description on the dataset page: https://huggingface.co/datasets/davanstrien/enc-brit-glm-ocr-v2-full.