# Apache-2.0 license

PhoBERT is a Vietnamese pre-trained language model optimized based on the RoBERTa architecture, suitable for Vietnamese text classification tasks.

Nomic-embed-text-v1.5 is a text embedding model developed by Nomic AI, based on the sentence-transformers library, focusing on sentence similarity tasks.

A Turkish large language model developed by WiroAI, fine-tuned based on Llama-3.1-8B-Instruct, specializing in Turkish text generation and dialogue tasks.

A profanity detection model trained on 200K text data points for identifying inappropriate language in text.

CodeLlama-7B model fine-tuned on the KStack dataset, optimized for Kotlin code generation

This model is an open-source model based on the Apache-2.0 license, supporting the Abkhaz language, primarily used for object detection tasks.

Korean instruction-tuned version based on Meta-Llama-3-8B-Instruct, specifically optimized for Korean understanding and generation capabilities

test-dare is a merged model based on Mistral-7B, combining multiple Mistral-7B variants through the DARE method to integrate the strengths of each model.

M2-BERT-128 is an 80-million-parameter retrieval model checkpoint proposed in the paper 'Benchmarking and Building Long-Context Retrieval Models with LoCo and M2-BERT'

GEITje-7B-chat is a Dutch conversation model based on the Mistral architecture, specifically optimized for chat and dialogue tasks.

Merlyn Education Safety 12B is a large language model focused on generating safe content for the education sector, developed by Merlyn Mind.

Japanese StableLM Instruct Gamma 7B is a Japanese large language model developed by Stability AI, based on the Mistral architecture, focusing on Japanese instruction understanding and generation tasks.

Based on the meta-llama/Llama-13b-hf large model, fine-tuned and optimized on a dataset of 650,000 instructions for Korean

Mistral 7B v0.1 is a 7-billion-parameter open-source large language model developed by Mistral AI, released under the Apache 2.0 license.

A sentence similarity calculation model based on the CompCSE dedicated dataset, suitable for English text processing.

A RoBERTa-based masked language model pre-trained on Japanese academic paper abstracts from CiNii Articles

This is a customized compact version of bert-base-multilingual-cased, specifically optimized for Turkish, capable of fully replicating the original model's representation output while ensuring no loss of original precision.

This is a customized streamlined version of bert-base-multilingual-cased, specifically optimized for Russian language processing while preserving the original model's representational capabilities.

This model is a Spanish Q&A model fine-tuned on the SQAC dataset based on roberta-base-bne

A compact multilingual BERT version trained on Wikipedia datasets, specifically optimized for German while maintaining the original model's representation output and accuracy.