FMRIPrep is a robust preprocessing pipeline for functional MRI (fMRI) data that automatically adapts to the idiosyncrasies of virtually any dataset, ensuring high-quality preprocessing with no manual intervention. It is designed to be analysis-agnostic, transparent, and easy to use, with a modular architecture that combines tools from existing neuroimaging packages. FMRIPrep produces comprehensive visual reports that facilitate quality control and transparency, allowing researchers to scrutinize preprocessing results and communicate methods effectively. The tool has been validated on a diverse set of fMRI data from 54 studies in the OpenfMRI repository, demonstrating its robustness and quality. FMRIPrep improves spatial accuracy by introducing less uncontrolled spatial smoothness than commonly used preprocessing tools. It also supports a wide range of output spaces, including subject-space and atlases for both volume-based and surface-based analyses. The tool is open-source, community-driven, and developed with high standards of software engineering. It supports execution via Docker and Singularity container platforms, ensuring reproducibility. FMRIPrep has been shown to produce high-quality results on a diverse set of input data, with validation phases demonstrating its reliability and effectiveness. The tool is particularly effective in reducing spatial smoothing and improving spatial precision, as demonstrated by comparisons with the FSL feat workflow. FMRIPrep's modular design allows for easy extension to support specific populations or new species by providing appropriate atlases. The tool is also robust in handling challenging data, such as from simultaneous MRI/electrocorticography (ECoG) studies. FMRIPrep's transparent and accessible development and reporting aims to better equip fMRI practitioners to perform reliable, reproducible, statistical analyses with a high-standard, consistent, and adaptive preprocessing instrument.FMRIPrep is a robust preprocessing pipeline for functional MRI (fMRI) data that automatically adapts to the idiosyncrasies of virtually any dataset, ensuring high-quality preprocessing with no manual intervention. It is designed to be analysis-agnostic, transparent, and easy to use, with a modular architecture that combines tools from existing neuroimaging packages. FMRIPrep produces comprehensive visual reports that facilitate quality control and transparency, allowing researchers to scrutinize preprocessing results and communicate methods effectively. The tool has been validated on a diverse set of fMRI data from 54 studies in the OpenfMRI repository, demonstrating its robustness and quality. FMRIPrep improves spatial accuracy by introducing less uncontrolled spatial smoothness than commonly used preprocessing tools. It also supports a wide range of output spaces, including subject-space and atlases for both volume-based and surface-based analyses. The tool is open-source, community-driven, and developed with high standards of software engineering. It supports execution via Docker and Singularity container platforms, ensuring reproducibility. FMRIPrep has been shown to produce high-quality results on a diverse set of input data, with validation phases demonstrating its reliability and effectiveness. The tool is particularly effective in reducing spatial smoothing and improving spatial precision, as demonstrated by comparisons with the FSL feat workflow. FMRIPrep's modular design allows for easy extension to support specific populations or new species by providing appropriate atlases. The tool is also robust in handling challenging data, such as from simultaneous MRI/electrocorticography (ECoG) studies. FMRIPrep's transparent and accessible development and reporting aims to better equip fMRI practitioners to perform reliable, reproducible, statistical analyses with a high-standard, consistent, and adaptive preprocessing instrument.