We demonstrate the capability of embedded mean eld theory (EMFT) within the linear-scaling density-functional theory code onetep, which enables DFT-in-DFT quantum embedding calculations on systems containing thousands of atoms at a fraction of the cost of a full calculation. We perform simulations on a wide range of systems from molecules to complex nanostructures to demonstrate the performance of our implementation with respect to accuracy and eciency. This work paves the way for the application of this class of quantum embedding method to large-scale systems that are beyond the reach of existing implementations.Another option is to enforce orthogonality manually to avoid the non-additive term ever appearing: this can be done either by construction, 37 or by projecting out the environment orbitals.3840 Each of these options has its own strengths and weaknesses, and some are more suitable than others for a given situation.Recently, another quantum embedding method that avoids these issues has been pro-