Local Large Language Models in R with Ollama

This tutorial introduces the use of local large language models (LLMs) in R via the Ollama platform and the ollamar package, covering installation and setup, the generate() and chat() functions, and prompt engineering for text analysis tasks including summarisation, classification, and question answering. It is aimed at researchers in linguistics and digital humanities who want to apply LLMs to their data without sending sensitive information to external APIs. This tutorial is part of the Language Technology and Data Analysis Laboratory (LADAL), a free, open-access research infrastructure at the University of Queensland. LADAL provides tutorials, tools, and courses for researchers working with language data. All materials are freely available at https://ladal.edu.au and are part of the Language Data Commons of Australia (LDaCA), funded by ARDC and NCRIS.