🦋 NimbleBox ChainFury

🦋 Build complex chat apps using LLMs in 4 clicks ⚡️ Try it out here. Used in production by chat.nbox.ai.

Read the Docs

The documentation page contains all the information on using chainfury and chainfury_server.

Installation

There are two separate packages built into this repository, first is chainfury which contains the fury-engine for running the DAGs and chainfury_server which contains the self hosted server for the GUI.

pip install chainfury
pip install chainfury_server

# to launch the server
python3 -m chainfury_server

Run Docker

Easiest way to run the server is to use docker. You can use the following command to run ChainFury:

docker build . -f Dockerfile -t chainfury:latest
docker run -p 8000:8000 chainfury:latest

To pass any env variables you can use the command:

docker run --env ENV_KEY=ENV_VALUE -p 8000:8000 chainfury:latest

Checkout all the:

• component environment variables here
• chainfury specific variables here
• chainfury_server specific variables here

From Source

Here's a breakdown of folder:

• chainfury/ contains the chainfury engine
• server/ contains the chainfury server
• client/ contains the frontend code for the GUI
• api_docs/ contains the documentation

To build the entire system from scratch follow these steps:

git clone https://github.com/NimbleBoxAI/ChainFury
cd ChainFury
python3 -m venv venv
source venv/bin/activate

You will need to have yarn installed to build the frontend and move it to the correct location on the server

sh stories/build_and_copy.sh

Once the static files are copied we can now proceed to install dependecies:

pip install setuptools
pip install -e .          # editable install the chainfury
cd server
pip install -e .          # editable install the chainfury_server

To start you can now do:

cd chainfury_server
python3 server.py

You can now visit localhost:8000 to see the GUI.

Contibutions

ChainFury is an open-source project used in production. We are open to contributions to the project in the form of features, infrastructure or documentation.

• Our issues page is kept up to date with bugs, improvements, and feature requests.

• If you're looking for help with your code, hop onto GitHub Discussions board or Discord, so that more people can benefit from it.

• Describing your issue: Try to provide as many details as possible. What exactly goes wrong? How is it failing? Is there an error? "XY doesn't work" usually isn't that helpful for tracking down problems. Always remember to include the code you ran and if possible, extract only the relevant parts and don't just dump your entire script. This will make it easier for us to reproduce the error.

• Sharing long blocks of code or logs: If you need to include long code, logs or tracebacks, you can wrap them in <details> and </details>. This collapses the content so it only becomes visible on click, making the issue easier to read and follow.