• Settings menu
• Toggle when to use memory and when not
• Better memory context handling

Release Notes

• Significantly smaller bundle/binary size
• Conversations now have memory (no more repeating yourself!)
• Faster load times

Note: This is a breaking change. When running Ollama locally, you must have the nomic-embed-text embedding model installed on your machine. It's very lightweight, so it won't consume much storage or RAM/VRAM. To install:

ollama pull nomic-embed-text

Also, I recommend updating your .zshrc or .bashrc by adding an alias which points to com.capable.localai which is the package name for Local Ai.

alias localai="com.capable.localai"

• 🤖 Local AI Integration: Connect to Ollama models running on your machine
• 💬 Persistent Chat History: SQLite database stores all conversations locally
• 🔄 Streaming Responses: Real-time AI responses with typing indicators
• 📂 Multiple Conversations: Create and manage multiple chat sessions
• 🎨 Modern UI: Clean, responsive interface with dark/light theme support
• 📱 Mobile-Friendly: Responsive design with collapsible sidebar
• 🚀 Fast Performance: Built with Vite for lightning-fast development and builds
• 📄 Paginated History: Efficient loading of chat history with scroll-based pagination
• 🔧 Model Selection: Choose from available Ollama models via dropdown