AI-Sweden-Models/gpt-sw3-20b

The GPT-Sw3 20B model by AI Sweden is a 20.9 billion parameter decoder-only transformer language model, pretrained on 320 billion tokens across Swedish, Norwegian, Danish, Icelandic, English, and programming code. Developed in collaboration with RISE and WASP WARA, it is designed for generating coherent text in five languages and four programming languages. This model is primarily intended for research and evaluation of large language model capabilities within the Nordic languages.