The University of Cape Town (UCT) is set to establish Africa’s largest GPU intensive compute cluster dedicated to artificial intelligence (AI) research, marking a shift in the continent’s digital research capabilities, it says.
The African Compute Initiative (ACI) forms part of the ‘AI for Development’ programme, a £58 million partnership between the UK government’s Foreign, Commonwealth and Development Office and Canada’s International Development Research Centre.
It aims to address long-standing infrastructure constraints that have limited Africa’s ability to develop and scale AI technologies locally.
The infrastructure will be housed in UCT’s upgraded High-Performance Computing (HPC) Data Centre and is expected to support workloads such as model training, fine-tuning, inference and large-scale simulations.
With the computational power of the data centre, researchers can tackle a wide range of AI and data-intensive projects, says UCT. These include climate modelling and environmental simulations, epidemiological and health forecasting models, natural language processing for African languages and astronomical simulations.
By combining advanced GPUs, high-capacity storage and high-speed networking within UCT’s data centre, the ACI will enable researchers across the continent to build, train and deploy AI systems locally, it adds.
Closing Africa’s compute gap
According to UCT, the initiative is positioned to shift African institutions from being consumers of global AI technologies, to active contributors at the frontier of innovation.
“African researchers have the ideas and the talent, but they have been held back by a lack of access to the computing power that AI development demands,” says Jonathan Shock, associate professor of applied mathematics at UCT and interim director of the UCT AI initiative.
“The African Compute Initiative changes that. It means researchers and students across Africa can work at the frontier of AI, not just consume what is built elsewhere.”
The cluster is scheduled to become operational within 12 months, targeting around 100 active users in its first year and scaling to 300 users across at least five institutions within three years, says UCT.
The ACI builds on UCT’s existing expertise through the Inter-University Institute for Data-Intensive Astronomy (IDIA), which operates the ilifu research cloud. The ilifu platform has supported over 1 000 researchers in fields such as astronomy and bioinformatics since 2015 and currently serves more than 600 active users.
The system runs on OpenStack and Ceph technologies, which will also underpin the ACI, reducing deployment risk and accelerating implementation. IDIA’s infrastructure was recognised in 2024 with the NSTF-South32 Award for contributions aligned to the fourth industrial revolution.
UCT says its existing HPC resources are increasingly oversubscribed, and the ACI is designed to address this capacity shortfall, while expanding technical expertise through additional hires and specialised training programmes.
Beyond UCT, the ACI will extend access to labs and research hubs across Africa. The initiative includes on-boarding programmes and training workshops, with a focus on early-career researchers and underrepresented groups, says the institution.
The project is complemented by a social science research component led by Annette Hübschle from UCT’s Global Risk Governance Programme. This work, supported by the Mozilla Foundation, will analyse access patterns, identify participation barriers and develop frameworks for equitable allocation of compute resources.
The ACI integrates environmental sustainability through a planned 180kWp solar photovoltaic installation on UCT’s upper campus. The renewables system is expected to generate between 220MWh and 240MWh of renewable energy annually, offsetting approximately 200 tonnes of CO₂ each year.
Share
