WillSmith is an open-source alternative to LangSmith, designed to provide a more cost-effective and flexible solution for monitoring, debugging, and optimizing applications based on large language models (LLMs). It supports easy deployment via Docker and features a sleek user interface.
This project provides a docker-compose.yaml
file that includes everything you need to run the app and MongoDB. Make sure you have Docker and Docker Compose installed.
First, set up your environment variables in a .env
file:
# .env
MONGODB_URI=mongodb://your-mongo-uri:27017/willsmith
Then, simply run:
docker-compose up -d
Docker Compose will start both the WillSmith app and the MongoDB instance. The app will be available at http://localhost:3000
.
Once the app is running, open your browser and visit http://localhost:3000
to access the UI and start managing your application data.
To integrate WillSmith with your LangChain project for chain call monitoring, just add the following environment variables to your .env
file:
# In your LangChain project's .env file
LANGCHAIN_ENDPOINT=http://localhost:3000 # The service URL after Docker starts
LANGCHAIN_TRACING_V2=True
LANGCHAIN_API_KEY=your_api_key # Any value, no key validation yet
LANGCHAIN_PROJECT=project_name # Your custom project name
With these settings, LangChain will automatically send chain call data to WillSmith, which will store it in MongoDB.
We have an exciting set of features in our development pipeline. Here’s what’s coming next:
Feel free to extend the project based on your needs. We welcome contributions from the community! 🙌