language_model_gateway

Overview

This project is a language model gateway that provides an OpenAI compatible API for language models. It is built using FastAPI and GraphQL.

Prerequisites

• Docker
• Docker Compose
• Make

Getting Started

To run the project locally, follow these steps:

1. Clone the repository:

   git clone <repository-url>
   cd <repository-directory>

2. Create docker.env file in the root of the project based on the docker.env.example. Update the keys for the functionality/providers you're planning on using. AWS_CREDENTIALS_PROFILE is the only one that is absolutely required to get going. Set this to the AWS profile you're part of e.g., admin_dev.

3. Set up the development environment:

   make devsetup

4. Start the Docker containers:

   make down; make up

Project Architecture

The project is structured into several main components:

• API: Built using FastAPI, it provides endpoints for interacting with the language models.
• GraphQL: Used for querying and mutating data.
• Routers: Define the routes for different functionalities like chat completions, image generation, etc.
• Managers: Handle the business logic for different functionalities.
• Utilities: Helper functions and classes used across the project.
• Tools: AI Agents that perform specific tasks like querying databases, calling APIs, etc.

Dependencies

The main libraries and tools used in this project are:

• FastAPI
• GraphQL
• Docker
• Docker Compose
• Make
• Pipenv
• Uvicorn
• Datadog

Usage Examples

Here are some sample API requests and responses:

Chat Completions

Request:

curl -X POST "http://localhost:5050/api/v1/chat/completions" -H "Content-Type: application/json" -d '{
  "model": "gpt-3.5-turbo",
  "messages": [{"role": "user", "content": "Hello, how are you?"}]
}'

Response:

{
  "id": "chatcmpl-123",
  "object": "chat.completion",
  "created": 1677652288,
  "model": "gpt-3.5-turbo",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "I am fine, thank you!"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 9,
    "completion_tokens": 7,
    "total_tokens": 16
  }
}

Running without OAuth

Just run the following commands to run OpenWebUI without OAuth:

make down; make up; make up-open-webui

Running with OAuth

Since the OpenWebUI uses Keycloak on both server side and browser side, you need to create a host mapping for keycloak.

On Macs, this can be done by adding an entry to /etc/hosts:

127.0.0.1   keycloak

Then run:

make down; make up; make up-open-webui-auth

Contributing

We welcome contributions to the project! To contribute, follow these steps:

1. Fork the repository.
2. Create a new branch for your feature or bugfix.
3. Make your changes and commit them with a clear message.
4. Push your changes to your fork.
5. Create a pull request to the main repository.

Reporting Issues

If you encounter any issues, please report them by creating a new issue in the repository's issue tracker.

Frequently Asked Questions (FAQs)

How do I set up the environment variables?

Create a docker.env file in the root of the project based on the docker.env.example file. Update the keys for the functionality/providers you're planning on using.

How do I run the tests?

You can run the tests using the following command:

make tests

How do I add a new AI Agent?

Refer to the add_new_agent.md file for detailed instructions on how to add a new AI Agent.

Troubleshooting Tips

• If you encounter issues with Docker, try restarting the Docker service.
• Ensure that all environment variables are set correctly.
• Check the logs for any error messages and stack traces.