3 ms·
TL;DR. All of the LLM models tested playing chess performed terribly bad against Stockfish engine except gpt-3.5-turbo-instruct, which is a closed OpenAI model
by teleforce 2y ago
TL;DR.
All of the LLM models tested playing chess performed terribly bad against Stockfish engine except gpt-3.5-turbo-instruct, which is a closed OpenAI model.