This project provides a fully-featured, responsive web interface for interacting with Ollama Large Language Models, targeting users who want a quick and easy way to get started with LLMs locally or offline. It offers a ChatGPT-like user experience with features like code highlighting, chat history, and model management directly from the UI.

How It Works

The interface is built with Next.js and leverages shadcn-ui for a clean, modern UI. It communicates with a running Ollama instance via its API, allowing users to download, delete, and switch between models. Chats are stored locally in localStorage, eliminating the need for a separate database and enhancing privacy and ease of use.

Quick Start & Requirements

• Docker: docker run -d -p 8080:3000 --add-host=host.docker.internal:host-gateway -e OLLAMA_URL=http://host.docker.internal:11434 --name nextjs-ollama-ui --restart always jakobhoeg/nextjs-ollama-ui:latest
• Local Install: Requires Node.js (18+) and npm. Clone repo, npm install, npm run dev.
• Prerequisites: Ollama must be downloaded and running.
• Docs: https://github.com/jakobhoeg/nextjs-ollama-llm-ui

Highlighted Details

• ChatGPT-inspired UI for familiar user experience.
• Local storage for chat history, no database required.
• Direct model management (download, delete, switch) within the UI.
• Code syntax highlighting and one-click code copying.
• Light/Dark mode support.

Maintenance & Community

This is a hobby project. Links to community resources are not provided in the README.

Licensing & Compatibility

The project does not explicitly state a license in the README.

Limitations & Caveats

The project is a hobby project, and a more complete experience may be found elsewhere. Upcoming features are listed but not yet implemented.