This is a fine-tuned version of the Helsinki-NLP/opus-mt-en-ber model, specifically designed for English to Atlas Tamazight translation. The model is based on the MarianMT architecture and trained with 486,000 synthetic sentence pairs, supporting translation tasks for Tamazight, Tashelhit, and Central Atlas Tamazight.

Natural Language ProcessingSafetensorsMultiple Languages

This is a pre-trained language model that merges Vistral-24B-Instruct and Cydonia-24B-v4.2.0 using the SLERP method, specifically optimized to support bilingual processing capabilities in Russian and English. The model intelligently combines the advantages of the two base models, enhancing English performance while maintaining Russian capabilities.

Natural Language ProcessingTransformersMultiple Languages

This is a quantized version of Ilya626's Cydonia_Vistral model in GGUF format. Multiple quantization types are generated using the imatrix option of llama.cpp, supporting efficient operation under different hardware configurations.

Natural Language ProcessingGguf

This project uses the mergekit and DARE TIES methods to merge the soob3123/amoral-gemma3-12B-v2 and IlyaGusev/saiga_gemma3_12b models, aiming to provide more powerful language processing capabilities.

Natural Language ProcessingTransformers

LLaVA-Saiga-8b is a vision-language model (VLM) developed based on the IlyaGusev/saiga_llama3_8b model, primarily optimized for Russian tasks while retaining English processing capabilities.

MultimodalTransformersMultiple Languages

Russian-optimized version based on Llama3 8B model, supports text generation tasks

Natural Language ProcessingGgufOther

A Russian chat assistant based on Llama-3 8B Instruct, specially trained to provide Russian dialogue support.

Natural Language ProcessingTransformersOther

This model is a fine-tuned version of IlyaGusev/rut5_base_sum_gazeta, primarily used for Russian text summarization tasks.

Natural Language ProcessingTransformers

This is the AWQ quantized version of the Saiga Mistral 7B model developed by Ilya Gusev. AWQ is an efficient, accurate, and fast low-bit weight quantization method that supports 4-bit quantization. This version aims to provide faster inference speed and lower resource consumption while maintaining comparable quality to the original model. It supports various inference tools and environments, including text-generation-webui, vLLM, Hugging Face TGI, and Transformers.

Natural Language ProcessingTransformers

7B-parameter Russian large language model based on Mistral architecture, supporting text generation tasks

Natural Language ProcessingGgufOther

ru_turbo_alpaca is a Russian text generation model, fine-tuned based on the Alpaca model, suitable for Russian text generation tasks.

Natural Language ProcessingTransformersOther

Russian news summarization model based on MBART architecture, optimized for Gazeta.ru news articles

Natural Language ProcessingTransformersOther

Russian toxicity text classifier based on rubert-base-cased-conversational model

Natural Language ProcessingTransformersOther

Russian news headline generation model based on RuBERT, specifically optimized for Telegram content

Natural Language ProcessingTransformersOther

An extractive summarization model based on rubert-base-cased, specifically optimized for Russian news texts

Natural Language ProcessingTransformersOther

This is a text processing model supporting Russian, primarily used for text generation and text completion tasks.

Natural Language ProcessingTransformersOther

Russian abstractive summarization model based on rugpt3medium_based_on_gpt2, specifically trained on the Gazeta dataset

Natural Language ProcessingTransformersOther

This model predicts whether a causal relationship exists between two news headlines, supporting 7 types of relation labels, applicable to English and Russian texts.

Natural Language ProcessingTransformersMultiple Languages

Russian news headline generation model based on rut5-base, optimized for the Telegram platform

Natural Language ProcessingTransformersOther

A Russian abstractive summarization model based on rut5-base, optimized for Russian news summarization tasks

Natural Language ProcessingTransformersOther