Kronos - AI Personal Assistant, is a Python-based artificial intelligence application that uses the Langchain library, streamlit, and the Groq Free API to generate high-quality text based on user input. This project aims to facilitate faster and more efficient content creation.
For the project to work well, the following requirements are necessary:
- OS: Windows or Linux (preferred)
- Minimum Memory: 2 GB
- Minimum Storage: 10 GB
- Python: >=3.12
- Docker
- Groq API Key
GROQ_API_KEY=<your_api_key>
To get all Python dependencies, you can cat the pyproject.toml or requirements.txt files and see all the required dependencies.
make build
: Build the docker image and containersmake destroy
: Remove the docker image and containers
For Windows system or if you prefer, you can run the following commands in the terminal:
# Build
docker build --rm --no-cache -f docker/Dockerfile -t chatbot:latest .
docker compose -f docker/docker-compose.yml up -d
# Destroy
docker compose -f docker/docker-compose.yml down
docker rmi chatbot
docker builder prune -f
After the installation, access the application via the URL http:\\localhost:8501
.
The AI Personal Assistant called Kronos, can be used to create a various styles and formats of text's, such as:
- Generating text summaries from PDF or CSV files.
- Creating creative text for stories and poems
- Help develop new skills and improve abilities
You can define some settings, such as:
- AI model: Choose an AI model to generate text according to your needs.
- Temperature: The parameter that controls the model's creativity.
- Max Tokens: Sets the maximum number of tokens that the model can use to generate text.
- Response language: Chooses the response language between Português-BR or English-US.
This way, it is possible to define the best settings for each need.
For example, you can use the AI Personal Assistant to generate a summary of a long article or help you understand an article based on a specific topic.
Below is an demo of application usage:
- joaoallmeida 😎
This project is licensed under the MIT License - See the LICENSE file for more details.