The original intention of establishing the GMET online service platform is to provide an online interactive platform that uses ERA5 reanalysis data to calculate the atmospheric pressure at GNSS stations and the weighted average temperature (Tm) of the atmosphere above them, in order to facilitate the inversion of PWV by relevant researchers. In the past few decades, ground-based GNSS technology has proven its superiority in atmospheric water vapor detection due to its high precision, high temporal resolution, low cost, and all-weather characteristics. With the continuous increase in the number of ground-based GNSS stations, the application of ground-based GNSS meteorology has received widespread attention. Temperature and pressure are essential elements in ground-based GNSS water vapor inversion, but most ground-based GNSS stations were not initially established for meteorological research and were not equipped with meteorological sensors to measure temperature and pressure. Many stations equipped with meteorological sensors also have unreliable measurement results due to sensor calibration and drift issues. The latest ERA5 reanalysis data released by ECWMF has a time and spatial resolution of 1 hour and 31 kilometers. According to our research evaluation, the additional error introduced by using ERA5 to invert PWV based on temperature and pressure is less than 1 mm, and it can maintain the daily variation of PWV well (Zhang et al., 2019). However, high-resolution NWM products like ERA5 have a very large amount of data and their download speed is not ideal in China. It is based on this consideration that we have started to establish this online platform, which downloads NWM products such as ERA5 uniformly in the cloud and provides users with an online interactive interface. Users only need to input simple information to obtain the required meteorological parameters. Users do not need to repeatedly download massive ERA5 products, nor do they need to learn how to use ERA5 data and parameter extraction methods. Based on the open-source ray tracing software RADATE provided by TUWien（ https://vmf.ge0.tuwien.ac.at/ ）We found that using ERA5 data ray tracing to obtain tropospheric delay has some improvement compared to using ERA Interim (the previous generation product of ERA5 and the data source of VMF3 projection) (Zhou) et a1.2020)。 Therefore, the platform also provides users with numerical products based on ERA5 projection at designated sites.