Note Subset of Lichess puzzles used in the paper "Can You Learn an Algorithm? Generalizing from Easy to Hard Problems with Recurrent Networks"

nsarrazin/chessformer

Text Generation • 0.2B • Updated Apr 30 • 286 • 9

Note 231M-parameter autoregressive model trained on Lichess games from January 2023

Note Chess chatbot that makes use of the Lichess openings dataset

Note Chess openings explorer built using the Lichess openings dataset

Note Data used in https://raw.sh/posts/chess_puzzles where the author attempts to teach chat models to solve chess puzzles

Note Dataset used to train the DiffuSearch model (https://arxiv.org/abs/2502.19805 / https://github.com/HKUNLP/DiffuSearch / https://lichess.org/@/diffusearchv0)

Note VLM benchmark that uses Lichess data

Note https://arxiv.org/abs/2510.23948 ChessQA: Evaluating Large Language Models for Chess Understanding