Access Shuttle AI's API via an easy-to-use Python wrapper. Dashboard: https://shuttleai.app Discord: https://discord.gg/shuttleai


Keywords
shuttleai, ai, gpt, claude, api, free, chatgpt, gpt-4, api-wrapper, chatbot, chatgpt-api, openai, pypi, python
License
MIT
Install
pip install shuttleai==4.1.7

Documentation

The Official Python Library for the ShuttleAI API

Build, Lint, Format pypi Downloads Downloads/Month Python Versions

The ShuttleAI Python library provides easy access to the ShuttleAI REST API for Python 3.9+ applications. It includes type definitions for all request parameters and response fields, offering both synchronous and asynchronous clients powered by httpx and aiohttp, respectively.

We prioritize performance optimizations across the library. Beyond using orjson for near-instant JSON processing, we implement various techniques to reduce overhead and enhance speed, ensuring efficient and swift API interactions. These optimizations include minimizing built-in library usage, leveraging reusable aiohttp client sessions, and incorporating several small adjustments to streamline operations.

Installation

pip install shuttleai

We recommend using --upgrade or -U to ensure you have the latest version of the library.

From Source

This client uses poetry as a dependency and virtual environment manager.

You can install poetry with

pip install poetry

poetry will set up a virtual environment and install dependencies with the following command:

poetry install

Getting Started

Below is a non-streamed sync and async example for the ShuttleAI API. You can find more examples in the examples/ directory.

Synchronous Client

from shuttleai import ShuttleAI

shuttleai = ShuttleAI()

response = shuttleai.chat.completions.create(
    model="shuttle-2-turbo",
    messages=[{"role": "user", "content": "Imagine an AI like no other, its name is ShuttleAI."}],
)

print(chunk.choices[0].message.content)

Asynchronous Client

import asyncio
from shuttleai import AsyncShuttleAI

async def main():
    shuttleai = AsyncShuttleAI()

    response = await shuttleai.chat.completions.create(
        model="shuttle-2-turbo",
        messages=[{"role": "user", "content": "Imagine an AI like no other, its name is ShuttleAI."}],
    )

    print(response.choices[0].message.content)

asyncio.run(main())

ShuttleAI CLI

Scroll down to the Scripts section for more information.

Run examples

You can run the examples in the examples/ directory using poetry run or by entering the virtual environment using poetry shell.

Using poetry run

cd examples
poetry run python chat_no_streaming.py

Using poetry shell

poetry shell
cd examples

>> python chat_no_streaming.py

API Key Setup

To use the ShuttleAI API, you need to have an API key. You can get a FREE API key by signing up at shuttleai.app and heading to the key management page.

After you have an API key, you can set it as an environment variable:

Windows

setx SHUTTLEAI_API_KEY "<your_api_key>"

Note

This will only work in the current terminal session. To set it permanently, you can use the setx command with the /m flag.

setx SHUTTLEAI_API_KEY "<your_api_key>" /m

macOS/Linux

export SHUTTLEAI_API_KEY=<your_api_key>

Contribution

We welcome and appreciate contributions to the ShuttleAI API Python SDK. Please see the contribution guide for more information. Benefits may apply! 😄

Scripts

Formatting/Checks

  • poetry run ruff check shuttleai - Check ruff
  • poetry run black shuttleai --diff --color - Check black
  • poetry run black shuttleai - Format code
  • poetry run mypy shuttleai - Check for type errors

Tools

  • poetry run clean - Clean up the project directory
  • poetry run key - Display your default API key (if set by environment variable)
  • poetry run contr - Display contributors

ShuttleAI CLI

  • poetry run shuttleai or shuttleai - Run the ShuttleAI CLI. Example of Chatbot Example response of Chatbot

Important

We support auto TAB completion of commands and model names! Just press TAB! Example of TAB of Chatbot