Pipeline from OpenWebUi to FastApi Server

[Blainstorm2015]

Hello, I am relatively new to OpenWebUI. Currently, I am working on a project where I want to use OpenWebUI as the front end and a FastAPI server as the backend. The goal of this project is to build an Agent System using CrewAI. The system will leverage data from my database to enrich responses with specific information on various topics.

My current challenge is that I have no idea how to create such a pipeline. To start, I would like my server to simply print out the messages typed into the OpenWebUI frontend. Unfortunately, I haven’t been able to find anyone who has worked on something similar (or maybe I’m just searching in the wrong places).

Has anyone tried something similar or could provide guidance?

[rajtilak-2020]

To create a pipeline where OpenWebUI interacts with a FastAPI server, here's a high-level approach to get you started:

Step 1: Set Up FastAPI

1. Install FastAPI and Uvicorn:

   pip install fastapi uvicorn

2. Basic FastAPI Server: Create a simple server that prints incoming messages.

   from fastapi import FastAPI, WebSocket
   import logging
   
   app = FastAPI()
   
   @app.websocket("/ws")
   async def websocket_endpoint(websocket: WebSocket):
       await websocket.accept()
       while True:
           message = await websocket.receive_text()
           logging.info(f"Received message: {message}")
           await websocket.send_text(f"Message received: {message}")

Step 2: Connect OpenWebUI to FastAPI

OpenWebUI likely interacts with the backend through WebSocket or API calls. To integrate:

1. Identify OpenWebUI's event hooks or functions that send messages.
2. Use a WebSocket client (or HTTP requests if OpenWebUI supports it) to send typed messages to the FastAPI server.

Step 3: Testing Communication

• Run the FastAPI server:

  uvicorn main:app --reload

• Open OpenWebUI and input a message. Ensure that it connects to the FastAPI WebSocket endpoint. Messages should appear in the FastAPI logs.

Step 4: Enhance the Pipeline

Once basic communication works:

1. Integrate the CrewAI Agent System.
2. Query your database based on received messages.
3. Craft responses using CrewAI enriched with database data.

Debugging and Searching Resources

If you encounter challenges:

1. Search GitHub for examples involving OpenWebUI and FastAPI integration.
2. Explore WebSocket tutorials as they form the backbone of real-time communication.