GPT-SW3: a foundational model for Swedish NLP

Purpose and goal:

The project investigates the possibility of using a large-scale Swedish generative language model to create a general solution for text processing tasks that can be used by many different stakeholders. The project is based on the large-scale generative language model GPT-SW3, which AI Sweden is developing in collaboration with RISE and WASP WARA Media & Language.

The project aims to provide GPT-SW3 via a user-friendly web-based interface and validate the use of the model across various text processing tasks with clear needs owners. Expected results and effects:

If this project succeeds, this will lead to a disruptive change and acceleration in the use of AI-driven Swedish language technology by drastically reducing the threshold for using advanced language models to solve text processing tasks. Stakeholders will no longer need to invest in costly processes of collecting training data for their specific needs, nor will they have to purchase and operate expensive hardware resources.

The results and resources produced within this project will be applicable in virtually all sectors and industries. Approach and implementation:

The project brings together need owners from the private and public sectors and research organisations. AI Sweden and RISE develop the API solution during the first part of the project. In the next step, the project focuses on different applications.

The work will take place iteratively: the API solution is validated and improved with the help of results from the various applications (use cases). The need owners are responsible for their respective applications with the support of AI Sweden and RISE.