 The Qinghai-Tibet Plateau is rich in renewable solar energy resources. To take advantage of this, China has developed a dual carbon strategy which includes developing a global horizontal irradiance, GHI, prediction model, suitable for Tibet. Clouds, aerosols, air molecules, water vapor, ozone, carbon dioxide and other components all affect the amount of solar radiation reaching the surface. For the descending solar shortwave radiation flux in Tibet, the attenuation effect of clouds is the most important factor. Previously, artificial intelligence, AI, models were used to build GHI prediction models, but these models need to be optimized for each region. This study established a set of AI prediction models suitable for Tibet based on ground-based solar shortwave radiation flux observation and cloud cover observation data from the Yang-Beijing area. It was found that using cloud cover as a model input variable can significantly improve the prediction accuracy, reducing the RMSE by over 20 percent. Additionally, when the forecast horizon is one hour, the RMSE of the Random Forest and Long Short-Term Memory models with a 10-minute step decrease by 46.1. This article was authored by Lei Xiaowu, Tian Luchin, Niemet Siren, and others. We are article.tv, links in the description below.