PyPI | Documentation | Discord
Towards Automated General Intelligence
LionAGI is a cutting-edge intelligent agent framework. It integrates data manipulation with advanced machine learning tools, such as Large Language Models (i.e. OpenAI's GPT).
- Designed for data-centric, production-level projects,
- dramatically lowers the barrier in creating intelligent, automated systems
- that can understand and interact meaningfully with large volumes of data.
Install LionAGI with pip:
pip install lionagiDownload the .env_template file, input your appropriate API_KEY, save the file, rename as .env and put in your project's root directory.
by default we use OPENAI_API_KEY.
- supports all models in OpenRouter (not released yet, you can have a preview in
main) - supports all
langchainandllamaindexsplitters, chunkers, and readers (not released yet) - here is the the latest stable, please
pip install lionagi==0.0.111if you find 0.0.112 buggy
- Create a production ready LLM application in hours.
- written in pure python, minimum dependency
aiohttp,python-dotenv,tiktoken,pydantic - Efficient and verstile data operations for reading, chunking, binning, writing, storing and managing data
- (Work In Progress) support for all
langchainandllamaindexsplitters, chunkers, and readers - Unified interface with any LLM provider, API or local
- Fast and concurrent API call with
configurable rate limit - (Work In Progress) support for hundreds of models both API and local
- Fast and concurrent API call with
LionAGI is designed to be asynchronous only, please check python official documentation on how async work: here
Notice:
- calling API with maximum throughput over large set of data with advanced models i.e. gpt-4 can get EXPENSIVE IN JUST SECONDS,
- please know what you are doing, and check the usage on OpenAI regularly
- default rate limits are set to be tier 1 of OpenAI model
gpt-4-1104-preview, please check the OpenAI usage limit documentation you can modify token rate parameters to fit different use cases. - if you would like to build from source, please download the latest release, main is under development and will be changed without notice
The following example shows how to use LionAGI's Session object to interact with gpt-4 model:
import lionagi as li
# define system messages, context and user instruction
system = "You are a helpful assistant designed to perform calculations."
instruction = {"Addition":"Add the two numbers together i.e. x+y"}
context = {"x": 10, "y": 5}# in interactive environment (.ipynb for example)
calculator = li.Session(system=system)
result = await calculator.initiate(instruction=instruction,
context=context,
model="gpt-4-1106-preview")
print(f"Calculation Result: {result}")# or otherwise, you can use
import asyncio
async def main():
calculator = li.Session(system=system)
result = await calculator.initiate(instruction=instruction,
context=context,
model="gpt-4-1106-preview")
print(f"Calculation Result: {result}")
if __name__ == "__main__":
asyncio.run(main())Visit our notebooks for our examples.
We encourage contributions to LionAGI and invite you to enrich its features and capabilities. Engage with us and other community members Join Our Discord
When referencing LionAGI in your projects or research, please cite:
@software{Li_LionAGI_2023,
author = {Haiyang Li},
month = {12},
year = {2023},
title = {LionAGI: Towards Automated General Intelligence},
url = {https://github.com/lion-agi/lionagi},
}Python 3.9 or higher.