Leaf Area Index (LAI) with an explicit biophysical meaning is a critical variable to characterize terrestrial ecosystems. Long-term global datasets of LAI have served as fundamental data support for monitoring vegetation dynamics and exploring its interactions with other Earth components. However, current LAI products face several limitations associated with spatiotemporal consistency. In this study, we employed the Back Propagation Neural Network (BPNN) and a data consolidation method to generate a new version of the half-month 1/12° Global Inventory Modeling and Mapping Studies (GIMMS) LAI product, i.e., GIMMS LAI4g, for the period 1982−2020. The significance of the GIMMS LAI4g was the use of the latest PKU GIMMS NDVI product and 3.6 million high-quality global Landsat LAI samples to remove the effects of satellite orbital drift and sensor degradation and to develop spatiotemporally consistent BPNN models. The results showed that the GIMMS LAI4g exhibited higher accuracy than its predecessor (GIMMS LAI3g) and two mainstream LAI products (Global LAnd Surface Satellite [GLASS] LAI and Long-term Global Mapping [GLOBMAP] LAI), with an R2 of 0.95, mean absolute error of 0.18 m2/m2, and mean absolute percentage error of 15% which meet the accuracy target proposed by the Global Climate Observation System. It outperformed other LAI products for most vegetation biomes in a majority area of the land. It efficiently eliminated the effects of satellite orbital drift and sensor degradation and presented a better temporal consistency before and after the year 2000 and a more reasonable global vegetation trend. The GIMMS LAI4g product could potentially facilitate mitigating the disagreements between studies of the long-term global vegetation changes and could also benefit the model development in Earth and environmental sciences.