Paita - Python AI Textual Assistant

Paita is textual assistant for your terminal that supports multiple AI Services and models.

Key Features

• Supports Multiple AI Services: Paita integrates with a variety of AI services through the LangChain library. If AI service is compatible with LangChain then it can be used also with Paita.
• Textual User Interface on your terminal: Paita is based on Textual and provides a sophisticated user interface right within your terminal, combining the complexity of a GUI with console simplicity.
• Cross-Platform Compatibility: Paita is compatible with Windows, macOS, and Linux systems across most terminals; if Python runs in your environment and Textual supports it, then Paita will work.
• Supports Retrieval-Augmented Generation (RAG): Paita supports local vectorstore (Chroma) and crawling web page content.

Supported AI Services

• OpenAI
• AWS Bedrock
• Ollama (local models)

Getting Started

Prerequisites

• Python 3.8.1+
• Access to AI Service and configured in terminal

Installation and running

Install using pip (or pipx)

pip install paita

Run and enjoy!

paita

Some keyboard shortcuts

Paita is textual ui application so using keyboard shortcuts is recommended:

• Use tab and shift+tab to navigate between input field, send-button and question/answer boxes
• While question/answer box is focus use enter to "focus-in" and esc to "focus-out"
• Use c to copy content from question/answer box
• Contextual keyboard shortcuts are shown at the bottom of the UI

Configuring AI Service(s) and model access

OpenAI

OpenAI usage requires valid api key in environment variable.

export OPENAI_API_KEY=<OpenAI API Key>

AWS Bedrock

Enable AI model access in AWS Bedrock. Configure aws credential access accordingly.

Ollama

Ollama enables running chat models locally.

Install ollama for operating system or use official (docker image)[https://hub.docker.com/r/ollama/ollama]

Once ollama installed pull desired model from a registry e.g.

ollama pull llama3

By default paita connects to local Ollama endpoint. Optionally you can configure endpoint url with env variable:

export OLLAMA_ENDPOINT=<protocol>://<ollama-host-address>:<ollama-host-port>

Feedback

• Issues
• Discussion