NASA and IBM have jointly developed and released Surya, an open-source AI model designed to forecast solar weather with enhanced precision. Surya is trained on nine years of high-resolution imagery data from NASA's Solar Dynamics Observatory (SDO), totaling approximately 218 terabytes. The model can predict solar flares and geomagnetic storms up to two hours in advance, which could help protect critical technology and infrastructure from solar storm impacts. Surya improves solar flare classification accuracy by 16% compared to previous methods and is available on the Hugging Face platform for researchers worldwide. The AI model aims to provide earlier warnings for space weather events, including Carrington-class solar storms and auroras, potentially aiding agencies such as NASA and NOAA in better forecasting and preparedness. The Surya model, a 366 million parameter transformer, represents a collaboration between NASA's IMPACT AI team and IBM Research, marking a significant advancement in heliophysics and space weather prediction.