You need to agree to share your contact information to access this dataset
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
by downloading this dataset, you agree with the dataset license
Log in or Sign Up to review the conditions and access this dataset content.
zapytaj scrape
the full public archive of zapytaj.onet.pl, a Polish Q&A site, captured before its shutdown on 2026-09-30.
from datasets import load_dataset
questions = load_dataset("tiagozip/zapytaj", "questions", split="train")
answers = load_dataset("tiagozip/zapytaj", "answers", split="train")
format
| config | rows | notes |
|---|---|---|
| questions | 22,683,384 | one row per question, keyed by id |
| answers | 64,762,874 | one row per answer, joins to questions.id via question_id |
| options | 9,312,255 | poll, inquiry and quiz choices, joins via question_id |
| comments | 10,394,031 | comments on questions and answers, joins via question_id |
questions carry the title, body, author, category, vote counts, and the question kind (textual, poll, inquiry or quiz). answers carry the body, author, votes, and flags like best and by_expert. timestamps are parsed. every table has a source column: api rows come from the site's API, scraped 2026-09-28 to 2026-10-01, and html rows come from an earlier HTML scrape, used only for questions the API pass did not return.
raw/zapytaj-raw-20260930.tar.zst has the unprocessed scraper output (JSONL) for both passes.
license
disclaimers
access can be revoked at any time and all data is provided as-is. this dataset contains personal information about real people, a lot of whom are not public figures. please be respectful. every question and answer in this dataset was publicly readable at the time it was fetched.
opt out by emailing legal@tiago.zip.
citation
@misc{zapytaj,
title = {zapytaj: a full archive of zapytaj.onet.pl},
author = {tiagozip},
year = {2026},
url = {https://huggingface.co/datasets/tiagozip/zapytaj}
}
- Downloads last month
- 50