Chat with the Qwen 3 LLM directly on device without the need for an internet connection (it even works in airplane mode.) Jade uses the new MLX API made available by apple to enable completely local LLM inference. Download your model from a choice of 4 different Qwen models and chat. The Jade flavor of models are Qwen 3 models, finetuned on a comprehensive set of:

Grammar, Logic, Rhetoric and Math books
Systems Programming and Administration books (Apple focused, from aarch64 and XNU to userland Objective-C)
Systems Programming Documentation and Manpages (Rust, Cargo, Helix, Nushell, thousands of manpage entries)
Systems Programming Code Repositories (aarch64 Assembly, Hyper, Ripgrep, and SQLite)