POSE: Phase I: Developing Ecosystem and Community for OpenGPU Hardware and Software

Award ID: 2550223

Graphical Processing Units (GPUs) are essential to modern computing. They power artificial intelligence, scientific discovery, and advanced data processing. Yet much of todays GPU ecosystem is based on closed-source hardware and software. This limits access for universities, small companies, and the broader research community. This project addresses that challenge by building an open, community-driven GPU ecosystem, OpenGPU. By opening access to GPU infrastructure, the project supports more transparent and reproducible research, expands educational opportunities, and lowers barriers to innovation in Artificial Intelligence (AI), high-performance computing, and novel hardware systems. This project scopes and prepares a sustainable open-source ecosystem for OpenGPU, building on the teams existing open-source GPU architecture and software framework. The project develops governance structures, contributor processes, documentation, training materials, and the community infrastructure needed for long-term growth, while also improving technical practices such as automated testing, continuous integration and deployment workflows, and broader platform support. The main outcome is expected to be a stronger organizational and technical foundation for a scalable OpenGPU ecosystem. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Program Director: Marlon Pierce