kotaemon

Build and use local RAG-based Question Answering (QA) applications.

kotaemon-demo.mp4

Source Code | Live Demo

User Guide | Developer Guide | Feedback

This project would like to appeal to both end users who want to do QA on their documents and developers who want to build their own QA pipeline.

• For end users:
  • A local Question Answering UI for RAG-based QA.
  • Supports LLM API providers (OpenAI, AzureOpenAI, Cohere, etc) and local LLMs (currently only GGUF format is supported via llama-cpp-python).
  • Easy installation scripts, no environment setup required.
• For developers:
  • A framework for building your own RAG-based QA pipeline.
  • See your RAG pipeline in action with the provided UI (built with Gradio).
  • Share your pipeline so that others can use it.

This repository is under active development. Feedback, issues, and PRs are highly appreciated. Your input is valuable as it helps us persuade our business guys to support open source.

Installation

For end users

This document is intended for developers. If you just want to install and use the app as it, please follow the User Guide.

For developers

# Create a environment
python -m venv kotaemon-env

# Activate the environment
source kotaemon-env/bin/activate

# Install the package
pip install git+https://github.com/Cinnamon/kotaemon.git

Creating your application

In order to create your own application, you need to prepare these files:

• flowsettings.py
• app.py
• .env (Optional)

flowsettings.py

This file contains the configuration of your application. You can use the example here as the starting point.

app.py

This file is where you create your Gradio app object. This can be as simple as:

from ktem.main import App

app = App()
demo = app.make()
demo.launch()

.env (Optional)

This file provides another way to configure your models and credentials.

Configure model via the .env file

Alternatively, you can configure the models via the .env file with the information needed to connect to the LLMs. This file is located in the folder of the application. If you don't see it, you can create one.

Currently, the following providers are supported:

OpenAI

In the .env file, set the OPENAI_API_KEY variable with your OpenAI API key in order to enable access to OpenAI's models. There are other variables that can be modified, please feel free to edit them to fit your case. Otherwise, the default parameter should work for most people.

OPENAI_API_BASE=https://api.openai.com/v1
OPENAI_API_KEY=<your OpenAI API key here>
OPENAI_CHAT_MODEL=gpt-3.5-turbo
OPENAI_EMBEDDINGS_MODEL=text-embedding-ada-002

Azure OpenAI

For OpenAI models via Azure platform, you need to provide your Azure endpoint and API key. Your might also need to provide your developments' name for the chat model and the embedding model depending on how you set up Azure development.

AZURE_OPENAI_ENDPOINT=
AZURE_OPENAI_API_KEY=
OPENAI_API_VERSION=2024-02-15-preview
AZURE_OPENAI_CHAT_DEPLOYMENT=gpt-35-turbo
AZURE_OPENAI_EMBEDDINGS_DEPLOYMENT=text-embedding-ada-002

Local models

• Pros:
• Privacy. Your documents will be stored and process locally.
• Choices. There are a wide range of LLMs in terms of size, domain, language to choose from.
• Cost. It's free.
• Cons:
• Quality. Local models are much smaller and thus have lower generative quality than paid APIs.
• Speed. Local models are deployed using your machine so the processing speed is limited by your hardware.

Find and download a LLM

You can search and download a LLM to be ran locally from the Hugging Face Hub. Currently, these model formats are supported:

• GGUF

You should choose a model whose size is less than your device's memory and should leave about 2 GB. For example, if you have 16 GB of RAM in total, of which 12 GB is available, then you should choose a model that takes up at most 10 GB of RAM. Bigger models tend to give better generation but also take more processing time.

Here are some recommendations and their size in memory:

• Qwen1.5-1.8B-Chat-GGUF: around 2 GB

Enable local models

To add a local model to the model pool, set the LOCAL_MODEL variable in the .env file to the path of the model file.

LOCAL_MODEL=<full path to your model file>

Here is how to get the full path of your model file:

• On Windows 11: right click the file and select Copy as Path.

Start your application

Simply run the following command:

python app.py

The app will be automatically launched in your browser.

Customize your application

Please refer to the Developer Guide for more details.