# Multi-source data pre-training

Minueza-32M-Base is a base model with 32 million parameters, fully trained on extensive English text corpora, suitable for text generation tasks.

Arabic language model trained on T5v1.1 small architecture, incorporating multiple Arabic datasets for training

RoBERTa model pre-trained on massive Hindi data, supporting masked language modeling tasks

A Finnish ConvBERT generator model pre-trained with Replaced Token Detection (RTD) objective, specialized for fill-mask tasks.

A 345 million parameter GPT-2 model pre-trained on massive Finnish text, excelling in Finnish text generation