Hi HN,

I built a chess engine that utilizes LLM for thinking, basically it will send the current state and memory to an LLM and let the AI make a LEGAL move.

We can actually now see how good LLMs are against other specifically trained AIs for chess like stockfish. Any feedback is appreciated.