Lumen is a code based on Yambo 4.1.3 that implements a new approach to study non-linear response in solids and nanostructures based on a real-time evolution of an effective Schödinger equation.

Notice that all the non-linear implementation of Lumen will be soon included in the official Yambo distribution, so if you are not in hurry to use it, just wait for the new version of Yambo.


  • 07/06/2019: Lumen is merged in Yambo 4.4.0.
    All the non-linear part of Lumen has been merged back in Yambo version 4.4.0 that will be released soon. Stay tuned on the Yambo website.

  • 05/10/2018: Lumen 1.3 is online.
    Version 1.3 of Lumen is online. In this new version interrupted runs can be analized by means of ypp. A new test for collisions is added to compare collision in g-space and Kohn-Sham space.

  • 16/01/2018: Lumen 1.2 is online.
    A new version of Lumen is available online. In this new version the real-time Bethe-Salpeter equation has been implemented in terms of collisions as described in the appendix of Phys. Rev. B 84, 245110(2011).

  • 22/11/2017: Lumen 1.1 is online and a new tutorail on Third Harmonic Generation is available
    A new version of Lumen is available online. This version corrects some small bugs and improve parallelization for the real-time Bethe-Sapter equation and on OpenMP.
    A new tutorial on calculation of Third-Harmonic Generation in bulk silicon with Lumen is available here.

  • 07/11/2017: Lumen 1.0 is online
    We are proud to annunce that lumen version 1.0 is available online.
    Here a short presentation of the code.

  • 27/06/2017: Polarized light
    Circular and linear polarized light can be used to study non-linear response, with input lumen -u -V nl.

  • 27/06/2017: Spin-orbit coupling
    Spin-orbit coupling can be included in non-linear calculations.

  • 03/05/2017: Real-time Bethe-Salpeter code
    We have released the real-time Bethe-Sapeter code with a tutorial. Notice that RT-BSE can be used only with a local exchange (NGOscll=1) or including few g-vectors due to its high computational cost, an improved version will be released soon.