chatgpt-command 0.1.0

Talk to ChatGPT from your terminal.

Quickstart

First you'll need to install the CLI:

pip install chatgpt-cli

Then, you'll need a OPENAI API key (you can get one here), and you'll need to export your API Key as an environment variable:

export OPENAI_API_KEY=<your api key>

Then you can start a conversation with ChatGPT:

chatgpt what is 2 + 2

You can also send multiline messages:

chatgpt '''
    Make this sentence more sophisticated:

    I like to eat pizza
    '''

Your messages in each terminal window are saved to ~/.chatgpt/{OS boot time}/{terminal pid}/chatlog.json. This means you can ask follow-up questions in a terminal window and start a new conversation by opening a new window.

Settings

Use a different model, like GPT-4

By default, the CLI uses the gpt-3.5-turbo model.

However, you can use a different model by passing the --model flag:

chatgpt --model=gpt-4 Complete this phrase: "Ravioli ravioli, give me the..."

You can also change the default model by setting the CHATGPT_CLI_MODEL environment variable:

export CHATGPT_CLI_MODEL=gpt-4

NOTE: The gpt-4 model is not yet available to everyone. You can join the wailist here.

Increase the request timeout

By default, the CLI will wait 120 seconds for a response from the API. You can increase this timeout by setting the CHATGPT_CLI_REQUEST_TIMEOUT_SECS environment variable:

export CHATGPT_CLI_REQUEST_TIMEOUT_SECS=600