Can pinch and count, know everything in the world. Learn about the past and the present, and talk about the various aspects of life.
This is an exploratory project, the purpose is to let the AI call the corresponding API based on the content entered by the user, sort out the final answer and return it to the user.
- Current city weather
- Connect to search engine
- current horoscope
- Query courier information
- Today's Hot News
- Popular movie and TV information
- generate image
- LangChain is a framework for developing applications powered by language models.
- Chainlit is an open-source Python package that makes it incredibly fast to build and share LLM apps.
This project uses Poetry to create a virtual environment, you can also use other tools, such as Anaconda.
BestAI uses Azure OpenAI by default, and you can modify it to OpenAI or other open source LLMs.
- Use the following command to install dependencies
> poetry install
- Enter the virtual environment
> poetry shell
- Run app
> chainlit run app.py -w
Remember that before running, you need to supplement the information in the environment variables. The .env.example file is modified to .env
In addition, you can also install dependencies through requirements.txt and use other virtual tools to run them.
The requirements.txt file is generated by the poetry tool:
> poetry export --without-hashes --format=requirements.txt > requirements.txt
BestAI provides a Docker deployment solution:
> docker build -t best-appchain-ai:v1 .
The BestAI demo has been deployed to Google Cloud:
> gcloud app deploy app.custom.yaml
For business cooperation, project exchange or other ideas, you can contact me by email. [email protected]
BestonAI is licensed under the MIT License. See the LICENSE file for more details.