# Knowledge-Intensive Tasks

Qwen-2.5-7B-base-RAG-RL is a large language model with 7B parameters trained from scratch on an unknown dataset, incorporating Retrieval-Augmented Generation (RAG) and Reinforcement Learning (RL) technologies.

GLM-4-32B-0414 is a large-scale language model developed by THUDM, with 32 billion parameters, suitable for various text generation tasks.

FANformer-1B is an autoregressive model enhanced with innovative periodic mechanisms for language modeling, featuring 1.1 billion non-embedding parameters and trained on 1 trillion tokens.

The Trillion-7B Preview is a multilingual large language model supporting English, Korean, Japanese, and Chinese. It achieves performance competitive with higher-computation models while maintaining lower computational requirements.

An 8B-parameter language model based on the Llama 3.1 architecture, created by merging multiple models with mergekit, capable of text generation

A 14B-parameter large language model focused on instruction-following tasks, supporting text generation capabilities.

A Korean-optimized version based on Google's Gemma-2-9b-it model, developed by Saltware, suitable for Korean natural language processing tasks.

The Jais series is a bilingual large language model optimized for Arabic, while also possessing strong English capabilities. The 30B-16K version has 30 billion parameters and supports a context length of 16,384 tokens.

This is an 8B-parameter model based on the Llama-3 architecture, merged using the mergekit tool, with strong text generation capabilities.

A large language model based on the Mistral architecture, trained on the Pile dataset, supporting English text processing