datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
wikitext_document_level
Wikitext Document Level
This is a modified version of https://huggingface.co/datasets/wikitext that returns Wiki pages instead of Wiki text line-by-line. The original readme is contained below.
Dataset Card for "wikitext"
Dataset Summary
The WikiText language modeling dataset is a collection of over 100 million tokens extracted from the set of verified
Good and Featured articles on Wikipedia. The dataset is available under the Creative Commons… See the full description on the dataset page: https://huggingface.co/datasets/EleutherAI/wikitext_document_level.VDR_MEGA_MultiDomain_DocRetrieval
Visual Document Retrieval Dataset
Overview
This dataset is designed for training visual document retrieval models. It combines multiple datasets from the VDR series, Colpali, and LlamaIndex to create the most comprehensive training resource for visual document retrieval tasks.
Dataset Structure
The dataset contains structured fields including unique identifiers with string lengths ranging from 45 to 50 characters, search query text with variable lengths between… See the full description on the dataset page: https://huggingface.co/datasets/racineai/VDR_MEGA_MultiDomain_DocRetrieval.Docmatix
Dataset Card for Docmatix
Dataset description
Docmatix is part of the Idefics3 release (stay tuned).
It is a massive dataset for Document Visual Question Answering that was used for the fine-tuning of the vision-language model Idefics3.
Load the dataset
To load the dataset, install the library datasets with pip install datasets. Then,
from datasets import load_dataset
ds = load_dataset("HuggingFaceM4/Docmatix")
If you want the dataset to link to the pdf files… See the full description on the dataset page: https://huggingface.co/datasets/HuggingFaceM4/Docmatix.DocVQA
Large-scale Multi-modality Models Evaluation Suite
Accelerating the development of large-scale multi-modality models (LMMs) with lmms-eval
🏠 Homepage | 📚 Documentation | 🤗 Huggingface Datasets
This Dataset
This is a formatted version of DocVQA. It is used in our lmms-eval pipeline to allow for one-click evaluations of large multi-modality models.
@article{mathew2020docvqa,
title={DocVQA: A Dataset for VQA on Document Images. CoRR abs/2007.00398 (2020)}… See the full description on the dataset page: https://huggingface.co/datasets/lmms-lab-encoder/DocVQA.DocHPLT
DocHPLT: A Massively Multilingual Document-Level Translation Dataset
Existing document-level machine translation resources are only available for a handful of languages, mostly high-resourced ones. To facilitate the training and evaluation of document-level translation and, more broadly, long-context modeling for global communities, we create DocHPLT, the largest publicly available document-level translation dataset to date. It contains 124 million aligned document pairs across 50… See the full description on the dataset page: https://huggingface.co/datasets/HPLT/DocHPLT.MMLongBench-DocDoclingMatix
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.fixtures_docvqaThis dataset includes 2 document images of the DocVQA dataset.
They are used for testing the LayoutLMv2FeatureExtractor + LayoutLMv2Processor inside the HuggingFace Transformers library.
More specifically, they are used in tests/test_feature_extraction_layoutlmv2.py and tests/test_processor_layoutlmv2.py.
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.DocHPLTv2DocLayNet-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.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.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.pixmo-docs
PixMo-Docs
We now recommend using CoSyn-400k and CoSyn-point over these
datasets. They are improved versions with more images categories and an improved generation pipeline.
PixMo-Docs is a collection of synthetic question-answer pairs about various kinds of computer-generated images, including charts, tables, diagrams, and documents.
The data was created by using the Claude large language model to generate code that can be executed to render an image,
and using GPT-4o mini to… See the full description on the dataset page: https://huggingface.co/datasets/allenai/pixmo-docs.transformers_image_docPubTables-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.docmath-eval-failures-200
DocMath-Eval Failures 200: Agent Benchmark & Leaderboard
A curated benchmark of 200 challenging financial math questions that leading AI models
failed to answer correctly, with comprehensive evaluation results from multiple AI agents.
Leaderboard
Evaluated on 2026-02-21 using LLM-as-Judge (Qwen QwQ-32B) for soft scoring.
Rank
Agent
Model
Exact Match
Judge: Exact
Judge: Approx
Judge: Total
Wrong
Avg Duration
Avg Tool Calls
1
TRAE Agent
Opus 4.5
98/200 (49.0%)
96… See the full description on the dataset page: https://huggingface.co/datasets/Ayushnangia/docmath-eval-failures-200.docvqa-single-page-questions
Dataset Card for DocVQA Dataset
Dataset Summary
DocVQA dataset is a document dataset introduced in Mathew et al. (2021) consisting of 50,000 questions defined on 12,000+ document images.
Please visit the challenge page (https://rrc.cvc.uab.es/?ch=17) and paper (https://arxiv.org/abs/2007.00398) for further information.
Usage
This dataset can be used with current releases of Hugging Face datasets library.
Here is an example using a custom collator to bundle… See the full description on the dataset page: https://huggingface.co/datasets/pixparse/docvqa-single-page-questions.DocumentVQAMMLongBench-docMP-DocVQAprotein-docs
Protein Documents (Parquet)
Structured text documents encoding protein residue sequences and 3D contact maps from AlphaFold Database v4 predicted structures, stored as Parquet files. Each row is one protein document with metadata.
Source structures: timodonnell/afdb-24M and timodonnell/afdb-1.6M
Document Schemes
Each subdirectory contains documents generated with a different scheme. All schemes share leakage-resistant train/val/test splits based on structural… See the full description on the dataset page: https://huggingface.co/datasets/timodonnell/protein-docs.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.doc-formats-parquet-1dr-saeid-ghezelbaash-entity-data
Dr. Saeed Ghezelbash Public Knowledge Graph
A public, physician-authored knowledge graph and multilingual retrieval dataset by Dr. Saeed Ghezelbash, a physician in Kermanshah, Iran. It connects physician identity, aesthetic medicine services, published question-answer content and cited evidence for entity resolution and evidence-grounded AI retrieval.
The canonical source is the official website and Dataset graph. This Hugging Face repository is its AI distribution. The… See the full description on the dataset page: https://huggingface.co/datasets/doctor-ghezelbaash/dr-saeid-ghezelbaash-entity-data.docvqa_test_subsampled_beirBEIR version of vidore/docvqa_test_subsampled.
doclaynet_processed
Dataset Card for "doclaynet_processed"
Clean version of DocLayNet ready for finetuning.
VisRAG-Ret-Test-MP-DocVQA
Dataset Description
This is a VQA dataset based on Industrial Documents from MP-DocVQA dataset from MP-DocVQA.
Load the dataset
from datasets import load_dataset
import csv
def load_beir_qrels(qrels_file):
qrels = {}
with open(qrels_file) as f:
tsvreader = csv.DictReader(f, delimiter="\t")
for row in tsvreader:
qid = row["query-id"]
pid = row["corpus-id"]
rel = int(row["score"])
if qid in qrels:… See the full description on the dataset page: https://huggingface.co/datasets/openbmb/VisRAG-Ret-Test-MP-DocVQA.docfinqa_h_goldenargilla_sdk_docs_raw_unstructured
Dataset info
This dataset contains documentation chunks from repositories (ADD REPOS).
Postprocessing
After some inspection, some chunks contain text too short to be meaningful, so we decided to remove those by removing chunks whose number of tokens (computed
with the same tokenizer of the model to be used for the embeddings) is lower or equal to the 5%:
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("BAAI/bge-base-en-v1.5")
df =… See the full description on the dataset page: https://huggingface.co/datasets/plaguss/argilla_sdk_docs_raw_unstructured.