omarkamali/wikipedia-monthly-next

🚀 Wikipedia Monthly

Last updated: December 25, 2025, 19:55 UTC

This repository provides monthly, multilingual dumps of Wikipedia, processed and prepared for easy use in NLP projects.

📊 Current Statistics

Metric	Current Export (December 2025)	All Exports (Total)
Languages	3	3
Articles	33.7K	33.7K

Usage

Load any language with a single line of code using 🤗 datasets. latest always refers to the most recent dump, while dated configs refer to specific dumps, starting from 20250901.

from datasets import load_dataset

# Load the English dataset from the latest dump
dataset = load_dataset("omarkamali/wikipedia-monthly", "latest.en", split="train", streaming=True)

🆕 Using Wikisets

Create your own Wikipedia-based dataset with efficient sampling, filtering, and more with Wikisets.

from wikisets import Wikiset, WikisetConfig

# Create a multi-language dataset
config = WikisetConfig(
    languages=[
        {"lang": "en", "size": 10000},      # 10k sample
        {"lang": "fr", "size": "50%"},      # 50% of French Wikipedia
        {"lang": "ar", "size": 0.1},        # 10% of Arabic Wikipedia
    ],
    date="20251101", # optional, defaults to latest
    seed=42
)

dataset = Wikiset.create(config)

# Access like any HuggingFace dataset
print(len(dataset))
print(dataset[0])

# {"id": "123", "url": "https://en.wikipedia.org/wiki/Example", "title": "Example", "text": "Example text"}

---

Data Fields

Each row in the dataset corresponds to a single Wikipedia article and contains the following fields:

• id: The unique Wikipedia page ID (string).
• url: The URL to the live article (string).
• title: The title of the article (string).
• text: The clean, plain text content of the article (string).

---

Why Use This Dataset?

1. Freshness: We run our pipeline monthly to capture the latest versions of all articles.
2. Clean & Ready: We handle the messy parts of parsing MediaWiki markup. You get clean plain text ready for use.
3. Easy Access: Load any language with a single line of code using 🤗 datasets.

Learn about the Wikipedia Monthly project in this blog post.

---

🌍 Language Subsets

This dataset is organized into configurations, one for each language dump. The table below lists all available subsets. For new languages, the article count will show as "Processing..." until the first full run is complete.

Language Code	Language	Date of last export	Articles	Size
ary	Ary	2025-12-01	10.9K	—
shi	Shi	2025-12-01	10.9K	—
zgh	Zgh	2025-12-01	11.9K	—

---

Dataset Creation Process

Our pipeline is designed for transparency and robustness:

1. Download: We fetch the latest pages-articles.xml.bz2 dump for each language directly from the official Wikimedia dumps server.
2. Filter: We stream the dump and process only main articles (namespace 0), filtering out user pages, talk pages, and other metadata.
3. Process: Using mwparserfromhell, we parse the MediaWiki syntax to extract clean, readable content.
4. Format: We generate a plain text representation by applying additional post-processing on the output from mwparserfromhell.
5. Upload: The resulting dataset is uploaded to the Hugging Face Hub with a configuration name corresponding to the dump date and language (e.g., 20250710.en).

Maintainer

Wikipedia Monthly is compiled, processed and published by Omar Kamali based on the official Wikipedia dumps.

License

Wikipedia Monthly is built on top of the incredible work by the Wikimedia Foundation and the open-source community. All content maintains the original CC-BY-SA-4.0 license.

Citation

If you find this dataset useful, please consider citing it:

@misc{wikipedia_kamali_2025,
    author       = { Omar Kamali and Omneity Labs },
    title        = { Wikipedia Monthly },
    year         = { 2025 },
    url          = { https://huggingface.co/datasets/omarkamali/wikipedia-monthly },
    doi          = { 10.57967/hf/6575 },
    publisher    = { Hugging Face }
}