#102 · Primary category： Foundation Models
kogpt
KakaoBrain KoGPT (Korean Generative Pre-trained Transformer)
Project last updated：01/30/24
GitHub Stars
1.0K
Forks
135
Contributors
11
License
Other
Why we included this project
KoGPT is a 6B-parameter Korean language model from KakaoBrain, released in float32 and float16 variants. You can load either one through the Hugging Face transformers API with a few lines of Python, and the repo includes a command-line inference script for trying prompts and sampling parameters before you wire the model into your own pipeline. The README publishes in-context few-shot results on Korean benchmarks such as NSMC, YNAT, and KLUE-STS, which gives a realistic sense of performance without running a full evaluation yourself. Hardware notes in the docs spell out the GPU memory needed to run it. It's a practical pretrained starting point for Korean text generation, classification, search, or summarization work, though the model is primarily Korean-focused.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
transformers
🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.
CLIP
CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image
MiniCPM-V
A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone
generative-models
Generative Models by Stability AI
unilm
Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities