SoilGrids1km is a global 3D soil information system that provides spatial predictions of various soil properties at a 1 km resolution. The system was developed using global soil profile databases and environmental covariates, including MODIS images, climate surfaces, lithological maps, and soil survey data. The predictions cover a range of soil properties such as organic carbon, pH, texture fractions, bulk density, cation-exchange capacity, coarse fragments, organic carbon stock, depth to bedrock, World Reference Base soil groups, and USDA Soil Taxonomy suborders. The accuracy of the predictions, assessed through 5-fold cross-validation, ranges from 23% to 51%. The main limitations of SoilGrids1km include weak relationships between soil properties and covariates, difficulty in obtaining comprehensive covariates, and low sampling density. However, the system is highly automated and flexible, allowing for the generation of increasingly accurate predictions as new data become available. SoilGrids1km is available for download under a Creative Commons non-commercial license.SoilGrids1km is a global 3D soil information system that provides spatial predictions of various soil properties at a 1 km resolution. The system was developed using global soil profile databases and environmental covariates, including MODIS images, climate surfaces, lithological maps, and soil survey data. The predictions cover a range of soil properties such as organic carbon, pH, texture fractions, bulk density, cation-exchange capacity, coarse fragments, organic carbon stock, depth to bedrock, World Reference Base soil groups, and USDA Soil Taxonomy suborders. The accuracy of the predictions, assessed through 5-fold cross-validation, ranges from 23% to 51%. The main limitations of SoilGrids1km include weak relationships between soil properties and covariates, difficulty in obtaining comprehensive covariates, and low sampling density. However, the system is highly automated and flexible, allowing for the generation of increasingly accurate predictions as new data become available. SoilGrids1km is available for download under a Creative Commons non-commercial license.