This project provides a user-friendly Swift-based graphical interface for interacting with local Large Language Models via Ollama.ai. It is designed for macOS users who want a seamless chat experience with their downloaded models, offering features like conversation management, model switching, and system prompt support.

How It Works

The application connects to a locally running Ollama instance, allowing users to select downloaded models, send prompts, and receive responses. It leverages Swift and SwiftUI for a native macOS experience, supporting features like dark mode, localization (English, Arabic), and a GUI for managing downloaded models.

Quick Start & Requirements

• Install Ollama: https://ollama.ai
• Open the Ollama Swift application.
• Download models via the "Manage Models" tab.
• Select a model from the dropdown on the main page to begin chatting.
• Requires macOS.

Highlighted Details

• Supports chatting with local LLMs.
• Allows changing models mid-conversation and restarting conversations.
• Features support for system prompts and multimodal models (image + text).
• GUI for downloading, deleting, and duplicating models.
• Localized interface with English and Arabic support.

Maintenance & Community

This appears to be a passion project with contributions from @HiRoS-neko, @Tooflex, and @chryoung. Feedback and contributions are welcomed via GitHub issues or email.

Licensing & Compatibility

The repository does not explicitly state a license in the provided README. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

The application requires a running Ollama instance to function. It is noted as a passion project, implying potential for imperfections and development delays. There is a standard macOS security warning for unverified developers that requires user intervention to bypass.