A lightweight, cross-platform chat application that connects to multiple LLM providers and stores user data locally in the browser. It requires an external paid API key to function.
Project overview
Offers a compact client (~5MB) for Linux, Windows, MacOS, and iOS that supports multiple AI providers alongside local data storage.
Project type
Model Runtime · Prompt Engineering
Use cases
Chat Assistants · Coding & Development
Deployment
Web app
License
MIT
Best for
General users seeking a cross-platform AI assistant that connects to multiple LLM providers while storing data locally in the browser.
Key capabilities
Provides a compact client (~5MB) running on Linux, Windows, MacOS, and iOS.
Stores all user data locally in the browser.
Compatible with multiple AI providers and self-deployed LLMs.
Allows creating, sharing and debugging chat tools with prompt templates (mask).
Easily preview, copy and share generated content/webpages through a separate window.
Supports network search, calculator, and other APIs.
Supports realtime chat capabilities.
Provides support for MCP when built with ENABLE_MCP=true.
Limitations and risks
Requires an external paid API key to function.
Getting started
Setup difficulty is easy. The documented first success path involves getting an OpenAI API Key, clicking Deploy with Vercel, and beginning use.
Alternatives and comparisons
Provides a simple way to get up and running with open models locally via CLI or REST API.
A CLI tool and Python library for interacting with large language models via remote APIs or local models.
Generate text, chat, quantize, and fine-tune Hugging Face LLMs natively on Apple silicon with MLX.