The main objective of this study is to investigate the impact of observation sampling rate on the PPP performance, in terms of positioning accuracy and convergence time, under different multi-GNSS combinations. For this purpose, daily observation datasets collected at ten IGS-MGEX (International GNSS Service-Multi-GNSS Experiment) stations during the 1-week period of January 6-12, 2019 with 1-s, 5-s,15-s, and 30-s sampling rates were processed under three different PPP processing modes, namely GPS-only, GPS/GLONASS and multi-GNSS. The results indicate that the use of high-rate observation sampling improves the PPP performance for all processing modes significantly. The contribution of increasing sampling rate, both for positioning error and convergence time, is more prominent in shorter periods, especially within the first 30 min for all processing modes. Finally, the multi-GNSS processing mode with the 1-s observation sampling rate has the best PPP performance and offers a considerable prospect for the GNSS applications.