kurumikz/telegram-corpus-russian-kazakh

📦 Telegram Corpus KAZ_RU Telegram Corpus KAZ_RU is a raw multilingual dataset of Telegram messages in Russian and Kazakh, collected and assembled by Kurumikz. It contains over 1.4 million lines of informal, user-generated text extracted from a large .txt dump. The corpus is intended for experimentation in text generation, language modeling, and dialogue research. 📊 Dataset Statistics Metric Value Description 📄 Lines 1,493,124 Total number of… See the full description on the dataset page: https://huggingface.co/datasets/kurumikz/telegram-corpus-russian-kazakh.