terminalgpt

Welcome to terminalGPT, the terminal-based ChatGPT personal assistant app! With terminalGPT, you can easily interact with the OpenAI GPT-3.5 and GPT-4 language models.

Whether you need help with a quick question or want to explore a complex topic, TerminalGPT is here to assist you. Simply enter your query and TerminalGPT will provide you with the best answer possible based on its extensive knowledge base.

Supported Models

• gpt-3.5-turbo
• gpt-4-turbo
• gpt-4o
• gpt-4o-mini
• o1
• o1-mini

Why?

Some advantages of using TerminalGPT:

• TerminalGPT's answers are tailored to your machine's operating system, distribution, and chip-set architecture.
• Doesn't use your conversation data for training the model.
• Your conversations are stored locally on your machine, so only you can access them.
• Efficient context management for super cheap api usage.
• It's highly available and can be used whenever you need it.
• You can use TerminalGPT with your IDE terminal, which means you won't have to constantly switch between your browser and your IDE when you have questions.
• Fast, reliable, lightweight.
• Quick one shot question mode.

Pre-requisites

• Python 3.6 or higher
• An OpenAI Account and API key.
  • Sign up at https://beta.openai.com/signup using email or Google/Microsoft account.
  • Go to https://beta.openai.com/account/api-keys or click on "View API keys" in the menu to get your API key.

Installation

Using pip

• Install the latest TerminalGPT with pip install.

pip install terminalgpt -U --user

or

pip3 install terminalgpt -U --user

Using pipx

• remove old installation that was installed with pip
  (do this only if you are migrating from a version that was installed with pip)

package_location=$(which terminalgpt)
rm "$package_location"

• install with pipx

pipx install terminalgpt==2.2.6 --force

Setup

• Now you have terminalgpt command available in your terminal. Run the following install command to configure the app.

terminalgpt install

• Enter your OpenAI API key when prompted and press enter.

• Choose one of the models below as the default model. it can be overridden with the -m --model flag later.

• Choose a printing style

That's it! You're ready to use TerminalGPT! You can now start a new conversation with terminalgpt new or load a previous conversation with terminalgpt load. Also you can reinstall with terminalgpt install or delete previous conversations with terminalgpt delete.

Usage

TL;DR

Usage: terminalgpt [OPTIONS] COMMAND [ARGS]...

  *~ TerminalGPT - Your Personal Terminal Assistant ~*

Options:
  --version                       Show the version and exit.
  -m, --model [gpt-3.5-turbo|gpt-4-turbo|gpt-4o|gpt-4o-mini|o1|o1-mini]
                                  Choose a model to use.  [default:
                                  gpt-4o-mini]
  -s, --style [markdown|plain]    Output style.  [default: plain]
  -t, --token-limit INTEGER       Set the token limit. this will override the
                                  default token limit for the chosen model.
  --help                          Show this message and exit.

Commands:
  delete    Choose a previous conversation to delete.
  install   Installing the OpenAI API key and setup some default settings.
  load      Choose a previous conversation to load.
  new       Start a new conversation.
  one-shot  One shot question answer.

New

Start a new conversation:

terminalgpt new

One-Shot (I love this feature)

One shot question to get a fast answer in the terminal.

terminalgpt one-shot "What is the meaning of life?"

Note: I recommend to make an alias, see at Recommended aliases.

Load

Load previous conversations:

terminalgpt load

Delete

Delete previous conversations:

terminalgpt delete

Recommended aliases

Zsh

echo alias tgpt="terminalgpt one-shot" >> ~/.zshrc
echo alias tgpto="terminalgpt one-shot" >> ~/.zshrc
echo alias tgptn="terminalgpt new" >> ~/.zshrc

Bash

echo alias tgpt="terminalgpt one-shot" >> ~/.bashrc
echo alias tgpto="terminalgpt one-shot" >> ~/.bashrc
echo alias tgptn="terminalgpt new" >> ~/.bashrc