_____ _ _                        ______ _____ _____ _____        ____    ___  
|  _  | | |                       | ___ \_   _/  ___/  __ \      / ___|  /   | 
| | | | | | __ _ _ __ ___   __ _  | |_/ / | | \ `--.| /  \/_   _/ /___  / /| | 
| | | | | |/ _` | '_ ` _ \ / _` | |    /  | |  `--. \ |   \ \ / / ___ \/ /_| | 
\ \_/ / | | (_| | | | | | | (_| | | |\ \ _| |_/\__/ / \__/\\ V /| \_/ |\___  | 
 \___/|_|_|\__,_|_| |_| |_|\__,_| \_| \_|\___/\____/ \____/ \_/ \_____/    |_/

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

# Building seems to be straight forward on the Ubuntu 24.04 LTS 

git clone https://github.com/ollama/ollama/ 
cd ollama
make -j8 


# Once build is finished run a server instance in one shell session 

./ollama serve 

# Run the DeepSeek model 
# https://ollama.com/library/deepseek-r1
# Will download about 4.7 Gigs first 

./ollama run deepseek-r1:7b 


# Dont expect super fast AI, but it works quite well