Frontier cloud models through OpenRouter, or entirely offline with LMStudio, llama.cpp, or Ollama. A pure-Swift engine enforces every rule; the model you've chosen plays the moves — with streaming commentary in the tone you pick.
ChessLLM keeps the rules honest in pure Swift, then hands the moves to whatever model you've pointed it at.
A pure-Swift engine enforces everything, for both sides.
Native SwiftUI throughout, in a familiar three-pane layout.
Switch tones mid-game and the next move's commentary uses the new one. Try them:
All four speak the same OpenAI-compatible chat-completions protocol. Switching is one click in Settings.
Frontier cloud models — Claude, GPT-4o, Llama 3.1, Gemini, and more. Bring your own key.
Point at the local server and pick a loaded model.
Point at llama-server and pick your loaded GGUF.
Point at the default port and pick any pulled model.
ChessLLM is free on the App Store. Bring a cloud key or a local model, choose a tone, and play.