BULaMU: An Open Foundation Model for Luganda

Uganda, colloquially referred to as the “pearl of Africa”, is home to over 50 million people, 56 tribes, and over 40 languages, most of which descend from the Bantu family. Luganda is the most popular language in Uganda, boasting over 20 million speakers around the globe. Around 70% of all the speakers of Luganda reside within the nation's borders, mostly occupying the Baganda area. This paper introduces BULaMU (Breakthrough in Utilization of Large Language Models in Uganda), the first Large Language Model that has been trained from scratch in Luganda. This model is both free and open weights and will be available on Huggingface to download for others to experiment with or adapt to their use case.