# Multilingual Understanding

Qwen3-30B-A3B-Base is the latest 30.5B parameter-scale Mixture of Experts (MoE) large language model in the Qwen series, supporting 119 languages and 32k context length.

A general-purpose text embedding model based on Qwen2-1.5B, supporting multilingual and long-text processing

Yi-1.5 is an upgraded version of the Yi model, demonstrating superior performance in programming, mathematics, reasoning, and instruction-following capabilities while maintaining excellent language understanding, commonsense reasoning, and reading comprehension abilities.

Re-uploaded version of Microsoft's multilingual MiniLM v2 model, optimized for use with the HuggingFace transformers library

mMiniLMv2-L12-H384 is a version of Microsoft's multilingual MiniLM v2 model, designed to provide a more convenient user experience through the HuggingFace transformers library.

mPLUG-Owl is a multilingual vision-language model that supports image understanding and multi-turn dialogue, developed based on the BLOOMZ-7B architecture.

A line-level document understanding model fine-tuned based on LiLT and DocLayNet dataset, supporting multilingual document layout analysis

CANINE-C is a character-based pre-trained Transformer model developed by Google, specifically designed for question answering tasks involving long text sequences.

XLM-RoBERTa is a multilingual pre-trained model based on the RoBERTa architecture, supporting 100 languages and suitable for cross-lingual understanding tasks.

XLM-RoBERTa-large is a multilingual pretrained language model based on the RoBERTa architecture, supporting various natural language processing tasks in multiple languages.

A Persian Q&A system fine-tuned based on the XLM-RoBERTa large model, trained on the PersianQA dataset, supporting Persian and multilingual Q&A tasks.

A multilingual pretrained model for Chinese and 7 minority languages in China

Vietnamese Q&A model fine-tuned from xlm-roberta-base, trained on UIT-Viquad_v2 dataset

XLM-RoBERTa is a multilingual pretrained model based on the RoBERTa architecture, supporting 100 languages and excelling in cross-lingual understanding tasks.

XLM-RoBERTa is a multilingual model pretrained on 2.5TB of filtered CommonCrawl data across 100 languages, using masked language modeling as the training objective.

A zero-shot classification model for Arabic, supporting natural language inference tasks

BERT is a pre-trained language model based on the Transformer architecture, capable of handling various natural language processing tasks in multiple languages.