# Korean Text Generation

Mamba-ko-2.8B is a Korean pre-trained model based on state space models, trained using the synthetically generated dataset korean_textbooks.

This is a Korean text generation model released under the Apache 2.0 license, specializing in Korean text-to-text generation tasks.

LLaMA-2-Korean Version is an advanced iteration of LLaMA 2, optimized for Korean text generation tasks through vocabulary expansion and additional Korean corpus pre-training.

KoAlpaca is the Korean version of the Stanford Alpaca model, combining the LLAMA architecture with Polyglot-ko technology, specifically optimized for Korean text generation tasks.

KoBART is a Korean encoder-decoder language model based on the BART architecture, trained with text infilling noise functions, supporting Korean text feature extraction and generation tasks.

KoBART is a Korean pretrained model based on the BART architecture, suitable for various Korean natural language processing tasks.

KoGPT2 is a Korean GPT-2 model developed by SKT-AI, based on the Transformer architecture, suitable for various Korean text generation tasks.