Ollama quick reference

Ollama commands and GPU requirements

Use these commands as a quick local AI starting point. Open the calculator when you need a hardware-specific result.

ModelCommandPlanning estimate
Llama 3.1 8B Instructollama run llama3.1:8b6.00 GB Q4 estimate
Llama 3.1 70B Instructollama run llama3.1:70b44 GB Q4 estimate
Qwen2.5 Coder 7Bollama run qwen2.5-coder:7b5.50 GB Q4 estimate
Qwen2.5 Coder 14Bollama run qwen2.5-coder:14b10.5 GB Q4 estimate
Qwen2.5 Coder 32Bollama run qwen2.5-coder:32b21 GB Q4 estimate
Qwen3 8Bollama run qwen3:8b6.00 GB Q4 estimate
DeepSeek R1 Distill Qwen 8Bollama run deepseek-r1:8b6.00 GB Q4 estimate
DeepSeek R1 Distill Qwen 14Bollama run deepseek-r1:14b10.5 GB Q4 estimate
DeepSeek R1 Distill Qwen 32Bollama run deepseek-r1:32b21 GB Q4 estimate
Gemma 3 4Bollama run gemma3:4b3.50 GB Q4 estimate
Gemma 3 12Bollama run gemma3:12b9.00 GB Q4 estimate
Gemma 3 27Bollama run gemma3:27b18 GB Q4 estimate
Mistral 7Bollama run mistral:7b5.50 GB Q4 estimate
Mixtral 8x7Bollama run mixtral:8x7b28 GB Q4 estimate
Phi-4 14Bollama run phi4:14b10.5 GB Q4 estimate