Dataset · tiagozip
Multi-Billion VK Public Data Scrap
tiagozip/vk
Regularly updated corpus of posts, comments, photos, videos, and profiles from VK in Russian and Ukrainian.
What's inside
A large-scale dump of public VK data (from 1 billion to 10 billion records) continuously collected since October 2026. Includes wall posts, comments, discussions, photos, videos, as well as community and user profile snapshots in Russian and Ukrainian. New posts and media are uploaded daily, while user and group data is updated weekly.
What you can do with it
- Pre-training and fine-tuning Russian and Ukrainian language models on natural speech
- Training text classifiers, comment moderation, and sentiment analysis in social networks
- Training multimodal models on text-image pairs and analyzing community graphs
- Tasks
- Text classification
- Languages
- Russian, Ukrainian
- Size
- 1B–10B rows
- Downloads this month
- 51.2K
- Likes
- 0
- Access
- on request on Hugging Face
How to load it
from datasets import load_dataset
ds = load_dataset("tiagozip/vk")