#ollama #chat #terminal #tui #ai

bin+lib chatti

Terminal-based chat application that interfaces with Ollama

1 unstable release

0.1.0 Oct 10, 2024

#275 in Machine learning

Download history 145/week @ 2024-10-06 30/week @ 2024-10-13 2/week @ 2024-10-20

177 downloads per month

MIT license

59KB
1K SLoC

Chatti

version Build Status Documentation codecov

Chatti is a terminal-based chat application that interfaces with Ollama, providing a unique way to interact with Ollama's language models directly from your command line.

Features

  • Support for various Ollama models
  • Configurable API endpoint and model parameters
  • Markdown rendering for chat responses

Prerequisites

Before you begin, ensure you have the following installed:

  • Rust (latest stable version)
  • Ollama (running and accessible)

Configuration

Chatti uses a configuration file located at ~/.config/chatti/config.toml. If this file doesn't exist, the application will create a default one on first run. You can edit this file to change the following settings:

api_endpoint = "http://localhost:11434/api/chat"
model = "llama3.2"
stream = true
temperature = 0.7
  • api_endpoint: The URL of your Ollama API endpoint
  • model: The Ollama model you want to use
  • stream: Whether to use streaming responses (recommended)
  • temperature: The temperature parameter for text generation (0.0 to 1.0)

Usage

To start the application, run:

cargo run

Once the application starts:

  • Type your message and press Enter to send it to the Ollama model
  • Use the up and down arrow keys to scroll through the chat history
  • Press 'q' to quit the application
  • Press '?' to display the help menu

Development

To run tests:

cargo test

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT

Dependencies

~25–39MB
~612K SLoC