Panchromatic modeling is one of the most powerful tools at our disposal to measure reliably the physical properties of galaxies across cosmic times. We present here an entirely new implementation in python of one such tool: CIGALE. Developed along three main design principles: simplicity, modularity, and efficiency, it has proven to be a versatile code that in addition to estimating the physical properties of galaxies (or regions within galaxies), can generate arbitrary sets of theoretical models or be used as a library to build other tools. Among its defining features, it is a truly panchromatic code ranging from the far-ultraviolet to the radio that takes into account numerous physical components (including active nuclei or synchrotron emission), that can fit non-photometric data, handle upper limits, determine photometric redshifts, and even build mock catalogs.