Install Ollama’s Local LLM on Windows and Power It with ShellGPT

Install Local Large Model (Windows)

Step 1: Install Ollama

Download the installer from https://ollama.com/download.

Step 2: Search on the Ollama website

Open the Ollama website and locate the desired model.

Step 3: Deploy in terminal

ollama run gemma2:2b

Linux Installation of ShellGPT

Step 1: Install ShellGPT

Run pip install shell-gpt --break-system-packages to install.

Step 2: Edit .sgptrc configuration file

Edit ~/.config/shell_gpt/.sgptrc and add the following options:

DEFAULT_MODEL=ollama/gemma2:9b
OPENAI_USE_FUNCTIONS=false
SHOW_FUNCTIONS_OUTPUT=false
API_BASE_URL=http://192.168.79.249:11434  # local server IP

Initialize ShellGPT

Provide any string as the initial API key.

sgpt

ShellGPT Feature Test

General function

Example: sgpt "你是谁" returns an introduction of the AI assistant.

Shell command generation

Example: sgpt -s "查看当前目录" suggests using pwd to display the current directory.

Code generation

Example:

sgpt -c "写一个shell脚本，要求获取cpu核数线程，内存大小，磁盘大小，网络带宽"

List chats

Use sgpt --list-chats to list all conversations.

Show chat info

Use sgpt --show-chat <path> to view a specific chat.

Start interactive REPL

Run sgpt --repl temp to enter REPL mode.