# Long Text Processing

Qwen3 is the latest generation of large language models in the Qwen series, offering a range of dense and Mixture of Experts (MoE) models. Based on extensive training, Qwen3 has achieved groundbreaking progress in reasoning, instruction following, agent capabilities, and multilingual support.

A fine-tuned LED model specifically for financial text summarization tasks, based on pszemraj/led-large-book-summary, developed by the Warwick Business School GEN AI GROUP 15 team

Qwen3 is the latest version of the Tongyi Qianwen series of large language models, offering a range of dense and mixture-of-experts (MoE) models. Based on large-scale training, Qwen3 has achieved breakthrough progress in reasoning, instruction following, agent capabilities, and multilingual support.

Qwen3 is the latest generation of large language models in the Qwen series, offering a range of dense and mixture-of-experts (MoE) models. Based on extensive training, Qwen3 has achieved breakthrough progress in reasoning, instruction following, agent capabilities, and multilingual support.

A medical reasoning model fine-tuned based on Qwen/Qwen2.5-7B-Instruct, excelling in multiple medical QA benchmarks

Qwen3-4B is the latest generation of the Qwen series large language model, featuring 4B parameters and supporting over 100 languages, with outstanding performance in reasoning, instruction following, and agent capabilities.

Qwen3-4B is the latest generation large language model in the Qwen series with 4B parameters, supporting over 100 languages and demonstrating exceptional reasoning, instruction following, and agent capabilities.

Chonky is a Transformer model capable of intelligently splitting text into meaningful semantic chunks, suitable for RAG systems.

A Japanese large language model based on the modernBERT-base architecture, supporting a maximum sequence length of 8192, trained on 3.4TB of Japanese corpus

Orpheus-3B-0.1-FT is a text-to-speech model based on a causal language model, supporting efficient quantization compression.

Chonky is a Transformer model that intelligently splits text into meaningful semantic chunks for RAG systems.

Shisa V2 is a bilingual (Japanese/English) general-purpose chat model trained by Shisa.AI, focusing on enhancing Japanese task capabilities while maintaining strong English proficiency.

A Japanese general-purpose reranking model built on ModernBERT-Ja, featuring top-tier performance and long-sequence processing capabilities

A Bible text relevance prediction model fine-tuned based on allenai/longformer-base-4096

A revolutionary open-source model capable of reordering long Arabic texts with exceptional efficiency and accuracy.

This is a cross-encoder model fine-tuned from ModernBERT-base, designed for calculating scores of text pairs, suitable for text reordering and semantic search tasks.

This is a cross-encoder model fine-tuned from ModernBERT-base for text re-ranking and semantic search tasks.

The fifth-generation Qwen2.5-YOYO model integrates features from multiple advanced models, optimizes the model merging formula, and supports a context length of 1 million tokens.

This is a cross-encoder model fine-tuned from ModernBERT-base for text re-ranking and semantic search tasks.

This is a cross-encoder model fine-tuned from ModernBERT-base, designed for calculating scores of text pairs, suitable for text re-ranking and semantic search tasks.

The EXAONE Deep series models excel in reasoning tasks such as mathematics and programming. This model is the 4-bit AWQ quantized version with 2.4 billion parameters

Kyro-n1.1 is an enhanced version of Kyro-n1, built upon Qwen2.5-7B-Instruct, with optimized reasoning capabilities, comprehension, and response accuracy.

DUO is a pretrained model based on the Transformers library, focusing on masked language modeling tasks and suitable for the field of natural language processing.

ARWKV-R1-1B5 is an early preview version of a 7-billion-parameter model based on RNN, trained through three-stage knowledge distillation from DeepSeek-R1-Distill-Qwen-1.5B, with a context length of 2k.

This is a GGUF format model converted from DISLab/SummLlama3.2-3B, primarily used for text summarization tasks.

A large language model fine-tuned for Japanese-to-English translation of light novels and web novels, supporting long-text translation up to 32K tokens

A model specifically designed for ultra-fast reranking, based on the JinaBERT architecture, supporting long text sequence processing (up to 8,192 tokens).

A legal domain sentence embedding model fine-tuned based on ModernBERT, supporting multi-level dimensional output, suitable for legal text similarity calculation and information retrieval tasks.

An 8-layer embedding model distilled from BAAI/bge-m3, achieving 2.5x speed improvement while maintaining retrieval performance

Eagle2 is a high-performance vision-language model family introduced by NVIDIA, focusing on enhancing the performance of open-source vision-language models through data strategies and training approaches. Eagle2-2B is the lightweight model in this series, achieving outstanding efficiency and speed while maintaining robust performance.

Eagle2-9B is the latest Vision-Language Model (VLM) released by NVIDIA, achieving a perfect balance between performance and inference speed. It is built on the Qwen2.5-7B-Instruct language model and the Siglip+ConvNext vision model, supporting multilingual and multimodal tasks.

GGUF format model file of OpenBuddy/openbuddy-deepseek-10b-v17.1-4k, supporting multiple languages, suitable for text generation tasks.

Jina Embeddings V2 Base Code is a transformer-based English sentence embedding model, specializing in feature extraction and similarity computation for code-related texts.

A large language model fine-tuned on multilingual datasets based on Qwen/Qwen2.5-7B, supporting 13 language processing

A Korean-English bilingual sentence embedding model optimized based on BAAI/bge-m3, supporting semantic text similarity, information retrieval, and other tasks

MiniCPM3-4B is the third-generation model in the MiniCPM series, with overall performance surpassing Phi-3.5-mini-Instruct and GPT-3.5-Turbo-0125, comparable to several recent 7B~9B-scale models.

A text ranking model fine-tuned with Korean data based on BAAI/bge-reranker-v2-m3

The Jais series are large language models specialized in Arabic processing while possessing strong English capabilities, with variants ranging from 590 million to 70 billion parameters.

The Jais family is a series of large bilingual language models specifically optimized for Arabic while maintaining strong English capabilities. This 6.7B parameter model is pre-trained from scratch and supports Arabic and English text generation.

A Spanish sentence embedding model fine-tuned based on BAAI/bge-m3, specifically optimized for the legal domain, suitable for semantic search and information retrieval tasks.