Leela available arguments:
-g [ --gtp ] Enable GTP mode.
--noponder Disable thinking on opponent's time.
-h [ --help ] Show commandline options.
-t [ --threads ] arg (=4) Number of threads to use.
-p [ --playouts ] arg Weaken engine by limiting the number of
playouts. Requires --noponder.
-b [ --lagbuffer ] arg (=100) Safety margin for time usage in centiseconds.
-l [ --logfile ] arg File to log input/output to.
-q [ --quiet ] Disable all diagnostic output.
-k [ --komiadjust ] Adjust komi one point in my disadvantage (for
territory scoring).
--nonets Disable use of neural networks.
--nobook Disable use of the fuseki library.
--gpu arg ID of the OpenCL device(s) to use (disables
autodetection).
--rowtiles arg (=5) Split up the board in # tiles.
Leela Zero available arguments:
-g [ --gtp ] Enable GTP mode.
--noponder Disable thinking on opponent's time.
-h [ --help ] Show commandline options.
-t [ --threads ] arg (=2) Number of threads to use.
-p [ --playouts ] arg Weaken engine by limiting the number of
playouts.Requires --noponder.
-v [ --visits ] arg Weaken engine by limiting the number of visits.
--timemanage arg (=auto) [auto|on|off] Enable extra time management
features.
auto = off when using -m, otherwise on
-b [ --lagbuffer ] arg (=100) Safety margin for time usage in centiseconds.
-r [ --resignpct ] arg (=-1) Resign when winrate is less than x%.
-1 uses 10% but scales for handicap.
-m [ --randomcnt ] arg (=0) Play more randomly the first x moves.
-n [ --noise ] Enable policy network randomization.
-s [ --seed ] arg Random number generation seed.
-d [ --dumbpass ] Don't use heuristics for smarter passing.
-w [ --weights ] arg File with network weights.
-l [ --logfile ] arg File to log input/output to.
-q [ --quiet ] Disable all diagnostic output.
--gpu arg ID of the OpenCL device(s) to use (disables
autodetection).
--full-tuner Try harder to find an optimal OpenCL tuning.
--tune-only Tune OpenCL only and then exit.