programming_ai_aid/Modelfiles

Mind the size of your GPU! Otherwise go for smaller model or a no GPU setting.

ollama list | grep deepseek

deepseek-r1:671b                 739e1b229ad7    404 GB    
deepseek-r1:70b                  0c1615a8ca32    42 GB     
deepseek-r1:32b                  38056bbcbb2d    19 GB     
deepseek-r1:14b                  ea35dfe18182    9.0 GB    
deepseek-r1:8b                   28f8fd6cdc67    4.9 GB    
deepseek-r1:7b                   0a8c26691023    4.7 GB    
deepseek-r1:1.5b                 a42b25d8c10a    1.1 GB    

CPU Versions (for a H100 with 94GB RAM)

ollama create code_ds1.5b -f Modelfile_Code_DS1.5b
ollama create code_ds7b -f Modelfile_Code_DS7b
ollama create code_ds8b -f Modelfile_Code_DS8b
ollama create code_ds14b -f Modelfile_Code_DS14b
ollama create code_ds32b -f Modelfile_Code_DS32b
ollama create code_ds70b -f Modelfile_Code_DS70b

CPU Versions (Set to use 24 cores)

ollama create code_ds1.5_cpu_24core -f Modelfile_Code_DS1.5b_CPU_24Core
ollama create code_ds7b_cpu_24core -f Modelfile_Code_DS7b_CPU_24Core
ollama create code_ds8b_cpu_24core -f Modelfile_Code_DS8b_CPU_24Core
ollama create code_ds14b_cpu_24core -f Modelfile_Code_DS14b_CPU_24Core
ollama create code_ds70b_cpu_24core -f Modelfile_Code_DS70b_CPU_24Core
ollama create code_ds32b_cpu_24core -f Modelfile_Code_DS32b_CPU_24Core
ollama create code_ds671b_cpu_24core -f Modelfile_Code_DS671b_CPU_24Core