Dataset · Hugging Face

ahmedheakl/docatlas_instruct

Name: docatlas_instruct
Creator: ahmedheakl
Keywords: task_categories:image-to-text, task_categories:document-question-answering, language:af, language:am, language:ar, language:az, language:ba, language:be, language:bg, language:bn, language:bs, language:ca, language:ce, language:ceb, language:ckb

DocAtlas: Multilingual Document Understanding Across 80+ Languages DocAtlas is a large-scale, high-fidelity multilingual OCR dataset covering 82 languages and 10 writing systems, built through model-free differential rendering. It provides precise structural annotations in a unified DocTag format encoding layout, text, and component types — without relying on any learned models for core annotation. This dataset powers the training of DocAtlas-DeepSeek, which… See the full description on the dataset page: https://huggingface.co/datasets/ahmedheakl/docatlas_instruct.

View on Hugging FaceReport an issue

Author / organization	ahmedheakl
Downloads	1.7K (as reported by Hugging Face)
Likes	1
Created	19 May 2026
Last updated at source	2 October 2026

Embed this entry

<iframe src="https://silytics.si/embed/item/68a4f80c-c971-485b-8407-6b385d2b2c73" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Popularity, last 3 days

1,716 now (+41 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.

More from ahmedheakl

ahmedheakl/daft-sm89-rdna3Dataset
ahmedheakl/daft-resultsDataset
ahmedheakl/resume-atlasDataset
ahmedheakl/wordocr_instruct_v4Dataset
ahmedheakl/arocrbench_hindawiDataset
ahmedheakl/ar_pixmocapqatrans_instructDataset

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.