# Wikipedia training

YugoGPT-Florida is a large language model based on Serbian, excelling in multiple evaluation benchmarks.

A small language model with only 16 million parameters, mainly used for debugging and testing, supporting English and Japanese.

A sentence-transformers model based on XLM-R, trained using the SimCSE method, which maps sentences and paragraphs into a 768-dimensional dense vector space, suitable for tasks such as clustering or semantic search.

This is a model based on sentence-transformers that can map sentences and paragraphs into a 768-dimensional dense vector space, suitable for tasks such as clustering or semantic search.

A SimCSE model based on mBERT, specifically trained for Thai language, used to generate 768-dimensional vector representations of sentences and paragraphs

Italian text ranking model based on MiniLMv2 architecture, with optimized embedding layers for Italian

A model that maps abstract sentence descriptions to matching sentences, trained on Wikipedia using a dual-encoder architecture.

A model that maps abstract sentence descriptions to matching sentences, trained on Wikipedia using a dual encoder architecture.

SHerbert large is an improved SentenceBERT model based on the Polish HerBERT, designed to generate semantically meaningful sentence embeddings and compare them using cosine similarity.

A language model fine-tuned for Guarani based on the multilingual BERT base model, supporting 104 languages including Guarani

A Danish GPT2-style model trained using the Flax CLM pipeline, based on the Danish portion of the wiki40b dataset.

This is a pre-trained Mongolian BERT model, trained on Mongolian Wikipedia and news datasets, supporting Mongolian text processing tasks.

A sentence transformer model based on m-Distil-BERT, trained using the SimCSE method, capable of mapping text to 768-dimensional vectors, suitable for semantic search and clustering tasks

A pre-trained BERT model for Galician (6 layers, case-sensitive). Trained on Wikipedia.

This is a Swedish GPT2-style model trained using the Flax CLM process, with training data from the Swedish portion of the wiki40b dataset.

A Korean pre-trained language model based on the BERT architecture, suitable for Korean text processing tasks.

A Thai sentence embedding model based on mBERT, trained using the SimCSE method on Thai Wikipedia data, capable of mapping text to 768-dimensional vectors

BERT model pre-trained on Thai Wikipedia text for POS tagging and dependency parsing

This is a customized streamlined version of bert-base-multilingual-cased, specifically optimized for Italian language processing while maintaining the original model's accuracy.

This is a compact version of distilbert-base-multilingual-cased, specifically designed for English and Vietnamese, maintaining the original model's accuracy.

A compact version customized from bert-base-multilingual-cased, focusing on English and Japanese processing while maintaining the original model's representational capabilities.

This is a lightweight version of distilbert-base-multilingual-cased, supporting English, French, Danish, Japanese, and Vietnamese processing while maintaining the original model's accuracy.

This is a lightweight version of distilbert-base-multilingual-cased, specifically optimized for Urdu, maintaining the accuracy of the original model.

This is a compact version of distilbert-base-multilingual-cased, specifically designed for bilingual tasks in English and Chinese, maintaining the accuracy of the original model.