Skip to main content

Getting Started

1. Install the latest release​

Homebrew (macOS and Linux)​

brew install gptscript-ai/tap/gptscript

Install Script (macOS and Linux):​

curl https://get.gptscript.ai/install.sh | sh

WinGet (Windows)​

winget install gptscript-ai.gptscript

Manually​

Download and install the archive for your platform and architecture from the releases page.

2. Get an API key from OpenAI.​

macOS and Linux​

export OPENAI_API_KEY="your-api-key"

Windows​

$env:OPENAI = 'your-api-key'

3. Run Hello World​

gptscript https://get.gptscript.ai/echo.gpt --input 'Hello, World!'

OUTPUT:

Hello, World!

The model used by default is gpt-4-turbo-preview and you must have access to that model in your OpenAI account.

4. Extra Credit: Examples and Run Debugging UI​

Clone examples and run debugging UI

git clone https://github.com/gptscript-ai/gptscript
cd gptscript/examples

# Run the debugging UI
gptscript --server