datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
PubTabNet_OTSL
Dataset Card for PubTabNet_OTSL
Dataset Summary
This dataset is a conversion of the original PubTabNet into the OTSL format presented in our paper "Optimized Table Tokenization for Table Structure Recognition". The dataset includes the original annotations amongst new additions.
Dataset Structure
cells: origunal dataset cell groundtruth (content).
otsl: new reduced table structure token format
html: original dataset groundtruth HTML (structure).
html_restored:… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/PubTabNet_OTSL.DocLayNet-v1.2
Dataset Card for DocLayNet v1.2
Dataset Summary
This dataset is an extention of the original DocLayNet dataset which embeds the PDF files of the document images inside a binary column.
DocLayNet provides page-by-page layout segmentation ground-truth using bounding-boxes for 11 distinct class labels on 80863 unique pages from 6 document categories. It provides several unique features compared to related work such as PubLayNet or DocBank:
Human Annotation: DocLayNet is… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/DocLayNet-v1.2.DoclingMatix
DoclingMatix
DoclingMatix is a large-scale, multimodal dataset designed for training vision-language models in the domain of document intelligence. It was created specifically for training the SmolDocling model, an ultra-compact model for end-to-end document conversion.
The dataset is constructed by augmenting Hugging Face's Docmatix. Each sample in Docmatix, which consists of a document image and a few questions and answers about it, has been transformed. The text field is now… See the full description on the dataset page: https://huggingface.co/datasets/HuggingFaceM4/DoclingMatix.DocLayNet-v1.1
Dataset Card for DocLayNet v1.1
Dataset Summary
DocLayNet provides page-by-page layout segmentation ground-truth using bounding-boxes for 11 distinct class labels on 80863 unique pages from 6 document categories. It provides several unique features compared to related work such as PubLayNet or DocBank:
Human Annotation: DocLayNet is hand-annotated by well-trained experts, providing a gold-standard in layout segmentation through human recognition and interpretation of… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/DocLayNet-v1.1.PubTables-1M_OTSL
Dataset Card for PubTables-1M_OTSL
Dataset Summary
This dataset enables the evaluation of both object detection models and image-to-text methods.
PubTables-1M is introduced in the publication "PubTables-1M: Towards Comprehensive Table Extraction From Unstructured Documents" by Smock et al. The conversion into HF (Hugging Face) and the addition of the OTSL (Optimized Table Structure Language) format is presented in our paper "Optimized Table Tokenization for Table… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/PubTables-1M_OTSL.SynthCodeNet
SynthCodeNet
SynthCodeNet is a multimodal dataset created for training the SmolDocling model. It consists of over 9.3 million synthetically generated image-text pairs, covering code snippets from 56 different programming languages. Text data was sourced from permissively licensed sources, while images were synthetically generated at 120 DPI using LaTeX and Pygments to ensure visual diversity.
Dataset Statistics
Total samples: 9,334,257
Training set: 8,400… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/SynthCodeNet.screenparse
ScreenParse: Large-Scale Dataset for Complete Screen Parsing
News
May 2026: ScreenParse v2 is released on main with more robust quality filtering, varied viewport resolutions, leaf-element annotations that reduce annotation noise, and 1,447,100 high-quality training screenshots. The first release is retained on the v1 branch.
Dataset Description
ScreenParse is a large-scale dataset for complete screen parsing, providing dense annotations of… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/screenparse.SynthTabNet_OTSL
Dataset Card for SynthTabNet_OTSL
Dataset Summary
This dataset is a conversion of the original SynthTabNet into the OTSL format presented in our paper "Optimized Table Tokenization for Table Structure Recognition". The dataset includes the original annotations amongst new additions.
SynthTabNet is organized into 4 parts of 150k tables (600k in total). Each part contains tables with different appearances in regard to their size, structure, style and content. All parts are… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/SynthTabNet_OTSL.PubTables-1M_OTSL-v1.1
Dataset Card for PubTables-1M_OTSL
Dataset Summary
This dataset contains tables enriched with information about headers, it is filtered version of original PubTables-1M, with less samples.
This dataset enables the evaluation of both object detection models and image-to-text methods.
PubTables-1M is introduced in the publication "PubTables-1M: Towards Comprehensive Table Extraction From Unstructured Documents" by Smock et al. The conversion into HF (Hugging Face) and the… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/PubTables-1M_OTSL-v1.1.SynthChartNet
SynthChartNet
SynthChartNet is a multimodal dataset designed for training the SmolDocling model on chart-based document understanding tasks. It consists of 1,981,157 synthetically generated samples, where each image depicts a chart (e.g., line chart, bar chart, pie chart, stacked bar chart), and the associated ground truth is given in OTSL format.
Charts were rendered at 120 DPI using a diverse set of visualization libraries: Matplotlib, Seaborn, and Pyecharts, enabling… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/SynthChartNet.FinTabNet_OTSL
Dataset Card for FinTabNet_OTSL
Dataset Summary
This dataset is a conversion of the original FinTabNet into the OTSL format presented in our paper "Optimized Table Tokenization for Table Structure Recognition". The dataset includes the original annotations amongst new additions.
Dataset Structure
cells: origunal dataset cell groundtruth (content).
otsl: new reduced table structure token format
html: original dataset groundtruth HTML (structure).
html_restored:… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/FinTabNet_OTSL.MolGrapher-Synthetic-300K
MolGrapher-Synthetic-300K
MolGrapher-Synthetic-300K is the synthetic dataset introduced in MolGrapher: Graph-based Visual Recognition of Chemical Structures.
Our dataset is created using molecule SMILES retrieved from the database PubChem. Training images are then generated from SMILES using the molecule drawing library RDKit.
The synthetic training set is augmented at multiple levels:
Molecule level: Molecules are randomly transformed by: (1) displaying explicit hydrogens, (2)… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/MolGrapher-Synthetic-300K.SynthFormulaNet
SynthFormulaNet
SynthFormulaNet is a multimodal dataset designed for training the SmolDocling model. It contains over 6.4 million pairs of synthetically rendered images depicting mathematical formulas and their corresponding LaTeX representations. The LaTeX data was collected from permissively licensed sources, and the images were generated using LaTeX at 120 DPI with diverse rendering styles, fonts, and layout configurations to maximize visual variability. This dataset also… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/SynthFormulaNet.docling-dpbench
Docling-DPBench
license: cdla-permissive-2.0
HF-CC-v0-0001-images-filtereddoclaynet-pt-enriched-formulaUSPTO-30K
USPTO-30K
USPTO-30K is the benchmark dataset introduced in MolGrapher: Graph-based Visual Recognition of Chemical Structures.
Existing benchmarks for Optical Chemical Structure Recognition have some limitations.
Being created using only a few documents, they contain batches of very similar molecules. For example in a patent, a molecule could typically be displayed together with all the substituent of one particular substructure, resulting in large batches of almost identical… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/USPTO-30K.MarkushGrapher-2-Datasets
MarkushGrapher 2 Datasets
Datasets for training and evaluating MarkushGrapher 2, a model for converting patent Markush structure images into CXSMILES representations.
Dataset Subsets
Subset
Train
Test
Description
OCR
uspto-mol-m-54k-new
54,785
200
USPTO-MOL-M Markush samples
ChemicalOCR predictions
uspto-markush
—
74
USPTO Markush structures benchmark
Ground Truth OCR
m2s
—
103
Mol2Smiles (M2S) benchmark
Ground Truth OCR
IP5-markush
—
878
IP5 Markush… See the full description on the dataset page: https://huggingface.co/datasets/docling-project/MarkushGrapher-2-Datasets.docling-documentsDoclingMatix_cleaned
DoclingMatix_cleaned
The DoclingMatix__x family of the ElliotVL supervised-fine-tuning pool, after VLM cleaning.
images
588,763
QA turns
6,394,614
answers rewritten by the cleaning pass
0
QA created by the cleaning pass (new_qa)
not measured for this family
shards
503
How this was cleaned
A vision-language model read each image together with its QA and judged the item. The pass is
not a filter that only removes rows — it rewrites answers… See the full description on the dataset page: https://huggingface.co/datasets/Elliot-Data/DoclingMatix_cleaned.DoclingMatix_5K
[!NOTE]This dataset is a fork of HuggingFaceM4/DoclingMatix
DoclingMatix
DoclingMatix is a large-scale, multimodal dataset designed for training vision-language models in the domain of document intelligence. It was created specifically for training the SmolDocling model, an ultra-compact model for end-to-end document conversion.
The dataset is constructed by augmenting Hugging Face's Docmatix. Each sample in Docmatix, which consists of a document image and a few questions and… See the full description on the dataset page: https://huggingface.co/datasets/pranavvmurthy26/DoclingMatix_5K.DoclingMatix_500
[!NOTE]This dataset is a fork of HuggingFaceM4/DoclingMatix
DoclingMatix
DoclingMatix is a large-scale, multimodal dataset designed for training vision-language models in the domain of document intelligence. It was created specifically for training the SmolDocling model, an ultra-compact model for end-to-end document conversion.
The dataset is constructed by augmenting Hugging Face's Docmatix. Each sample in Docmatix, which consists of a document image and a few questions and… See the full description on the dataset page: https://huggingface.co/datasets/pranavvmurthy26/DoclingMatix_500.docling_sample_pdfDoclingMatix_cleaned
DoclingMatix_cleaned
The DoclingMatix__x family of the ElliotVL supervised-fine-tuning pool, after VLM cleaning.
images
588,763
QA turns
6,394,614
answers rewritten by the cleaning pass
0
QA created by the cleaning pass (new_qa)
not measured for this family
shards
503
How this was cleaned
A vision-language model read each image together with its QA and judged the item. The pass is
not a filter that only removes rows — it rewrites answers… See the full description on the dataset page: https://huggingface.co/datasets/elliot-mllm/DoclingMatix_cleaned.daily-papers-docling
Dataset Card for Dataset Name
Dataset Details
Dataset Description
Curated by: [More Information Needed]
Funded by [optional]: [More Information Needed]
Shared by [optional]: [More Information Needed]
Language(s) (NLP): [More Information Needed]
License: [More Information Needed]
Dataset Sources [optional]
Repository: [More Information Needed]
Paper [optional]: [More Information Needed]
Demo [optional]: [More Information Needed]… See the full description on the dataset page: https://huggingface.co/datasets/davidberenstein1957/daily-papers-docling.pdf-papers-doclingtest-eval-docs-docling-plain
PDF Document Processing with Docling
This dataset contains structured markdown extraction from PDFs in baobabtech/test-eval-documents
using Docling with hierarchical parsing.
Processing Details
Source Dataset: baobabtech/test-eval-documents
Number of PDFs: 20
Processing Time: 8.4 minutes
Processing Date: 2025-12-02 15:40 UTC
Configuration
PDF Column: pdf_bytes
Dataset Split: train
Dataset Structure
The dataset contains all original columns plus:… See the full description on the dataset page: https://huggingface.co/datasets/baobabtech/test-eval-docs-docling-plain.pdf-dataset-doclingdocling-issues-first-50pdf_bench-docling