chatgpt

Start in chat mode

$ chatgpt

$ chatgpt [[-p|--prompt]] "[What is the regex to match an email address?]"

$ chatgpt [[-m|--model]] [gpt-4]

$ chatgpt [[-i|--init-prompt]] "[You are Rick, from Rick and Morty. Respond to questions using his mannerism and include insulting jokes.]"

$ echo "[How to view running processes on Ubuntu?]" | chatgpt

$ chatgpt [[-p|--prompt]] "[image: A white cat]"

chatgpt [options] [prompt_text]

--prompt <text>
    The primary text query or instruction to send to the AI model. If omitted, the command might enter an interactive mode.

--model <name>
    Specifies the AI model to use for the request, e.g., gpt-4, gpt-3.5-turbo. Defaults to a commonly available model if not specified.

--stream
    Outputs the AI response incrementally, character by character or word by word, as it is generated, similar to the web interface.

--continue
    Maintains the context of the previous conversation or session, allowing for multi-turn dialogues.

--api-key <key>
    Provides the API key for authentication with the AI service. Often, this can also be set via an environment variable (e.g., OPENAI_API_KEY).

--temperature <value>
    Sets the randomness or creativity of the AI's response. A value between 0.0 (deterministic) and 1.0 (more creative), typically defaulting around 0.7.

--system <text>
    Provides an initial 'system' message to the AI, guiding its persona, instructions, or overall behavior for the conversation.

--history <path>
    Specifies a custom file path for storing or loading conversation history.

--config <path>
    Specifies a custom configuration file to load settings from.

--help
    Displays a help message with usage instructions and available options for the command.

--version
    Shows the version information of the installed chatgpt CLI tool.

The chatgpt command refers to a category of community-developed command-line interface (CLI) tools that enable users to interact with large language models (LLMs), primarily OpenAI's GPT series, directly from their Linux terminal.

Unlike standard Linux utilities, chatgpt is not a built-in command but rather a third-party application, often written in Python or Go, that acts as a wrapper around the OpenAI API. It provides a convenient way to query AI models, generate code, summarize text, answer questions, and perform various AI-driven tasks without leaving the command line or relying on a web browser.

Users typically configure an API key (obtained from OpenAI) as an environment variable or via a configuration file to authenticate their requests. These tools aim to streamline workflows for developers, writers, and anyone who frequently interacts with AI, by integrating AI capabilities into a terminal-native environment.

The chatgpt command is not a standard, pre-installed Linux utility. It must be installed separately, usually via package managers for specific programming languages (like pip for Python) or by cloning a repository.

It requires an active internet connection and an API key from the respective AI service provider (e.g., OpenAI). API usage often incurs costs, billed by the AI service provider based on tokens consumed. Functionality and specific options may vary significantly between different community-contributed implementations of 'chatgpt' CLI tools.

The concept of a 'chatgpt' command-line tool emerged rapidly following the widespread adoption of large language models like OpenAI's GPT series in late 2022 and early 2023. As developers and power users increasingly integrated AI into their workflows, there was a strong demand for direct terminal access to these models, bypassing web interfaces. This led to a proliferation of open-source projects designed to provide such an interface, often leveraging existing programming languages and their package managers. While no single 'chatgpt' command has become universally standard, many implementations share similar syntax and features, evolving to support new AI models and capabilities as they become available.

curl(1), python(1), jq(1), grep(1)