Hello, Looking at the manuals, I saw that the function Py_NewInterpreter(), "simulates" the creation of a new thread, so that different scripts would be interpreted "pratically" in paralell by interpreting some bytecodes of each one from time to time. There is even a method to say how many bytecodes to interpret in one script (default=10) before changing to interpret the next script. I tryed to test this feature from C calls, by playing with thread states, interpreter states, run_string, etc, but with no success... What I would like to do is to have n scripts ( char *script[n]; ), and be able to run them in the same application loop, having something like : while ( true ) { for ( int i=0; i<n; i++ ) { restore_python_thread_state ( i ); // would call python functions to restore internal states, etc. run_some_bytecodes_of_python_script ( script[i] ); // would not block my loop, would not run all the script ! } .... } So, first of all, my question is : Is this possible to do ? If so, can someone explain me, or send me some example code please. Thank you all, Marcelo K. LIG - EPFL