diff --git a/README.md b/README.md
index b2853c1c09b2f41a358b7abb13cf7c6e6eb9721f..42d26e46f23967ba66f56cb25aec49475d212ba1 100644
--- a/README.md
+++ b/README.md
@@ -87,7 +87,7 @@ All of you has already received a personal USERNAME (`a08trbXY`) and a password.
 Once you are on `leonardo` login, submit an interactive job to get a compute node:
 
 ```bash
-srun -N 1 --ntasks-per-node=1 -A tra25_PythSci -p boost_usr_prod --reservation s_tra_py2 -t 02:00:00 --pty /bin/bash
+srun -N 1 --ntasks-per-node=1 -A tra25_PythSci -p boost_usr_prod --reservation s_tra_py3 -t 02:00:00 --pty /bin/bash
 ```
 The name of compute node will appear on the prompt once you got it (i.e.: `lrdn1342`):
 
@@ -104,7 +104,7 @@ All python packages will be available in the module, you don't need to create a
 
 **NB**: If more than one threads and a gpu are needed, it is necessary to request these resources to SLURM scheduler:
 ```bash
-srun -N 1 --ntasks-per-node=1 --cpus-per-task=4 --gres=gpu:1 -A tra25_PythSci -p boost_usr_prod --reservation s_tra_py2 -t 02:00:00 --pty /bin/bash
+srun -N 1 --ntasks-per-node=1 --cpus-per-task=4 --gres=gpu:1 -A tra25_PythSci -p boost_usr_prod --reservation s_tra_py3 -t 02:00:00 --pty /bin/bash
 ```
 
 **NB**: to access a compute node, you must exchange the ssh keys between compute and login nodes.