{"success":true,"data":{"author":"Google DeepMind (Anian Ruoss, Gregoire Deletang et al.)","bucketAccessType":"public","bucketListable":true,"bucketName":"searchless_chess","bucketPrefix":"data","category":"Machine Learning & AI","cloudProvider":"gcp","createdAt":"2026-11-21T20:49:29.591Z","description":"ChessBench is a benchmark of 10 million chess games with legal moves and value annotations from the Stockfish 16 engine, about 15 billion data points, released with the NeurIPS 2024 paper on amortized planning with large-scale transformers (searchless chess). Board states come from Lichess games. The data/ folder has train/ and test/ splits in .bag files (a simple record format read by the repository's bagz reader): action_value (win probability for every legal move of a position; the training set is sharded into 2,148 files, about 1.1 TB), behavioral_cloning (the best move per position, 34 GB train) and state_value (win probability per position, 36 GB train), plus puzzles.csv (Lichess puzzles with Elo ratings used for evaluation) and eco_openings.pgn. Positions are FEN strings and moves are UCI. Used to train transformer chess policies without search (a 270M-parameter model reached Lichess blitz Elo 2895), for planning and distillation research. License: CC BY 4.0, with the Lichess-derived portions CC0. Cite Ruoss, Deletang, Medapati et al., Amortized Planning with Large-Scale Transformers: A Case Study on Chess, NeurIPS 2024 (arXiv:2402.04494).","displayTitle":"ChessBench: 10 Million Chess Games with Stockfish 16 Action-Value, State-Value and Best-Move Annotations (Google DeepMind)","fuseMountBrowsable":true,"id":"251a55ab-5d9d-57e2-94f1-3a8e45b143d6","license":"CC-BY-4.0","licenseEvidenceUrl":"https://github.com/google-deepmind/searchless_chess","objectCountEstimate":2,"probeStatus":"ok","referenceUrl":"https://arxiv.org/abs/2402.04494","region":"us","schemaCard":{"accessHints":[{"language":"python-gcsfs","snippet":"import gcsfs\nfs = gcsfs.GCSFileSystem(token='anon')\nfs.ls('searchless_chess/data/', detail=False)[:10]"},{"language":"python-storage","snippet":"from google.cloud import storage\nclient = storage.Client.create_anonymous_client()\nlist(client.bucket('searchless_chess').list_blobs(prefix='data/', max_results=10))"}],"caveats":["Files are binary — use a domain-specific parser rather than text I/O."],"format":"binary","generatedAt":"2026-11-21T20:49:47.791Z","layout":{"example":"data/test/action_value_data.bag","pattern":"{id}/{seg1}/{filename}"},"mountPath":"/mnt/gcs-archive/data","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"data/test/action_value_data.bag","truncatedAt":512},{"bytes":"58/ppk2r2/2p1p2p/4ppb1/8/2P4P/PP2RPP1/3R2K1 b - - 5 29e5e4+8/8/P7/1N1p2Bp/8/2P3P1/4k2P/5RK1 b - - 0 40h5h4Br2q1rk1/pppnppbp/3p1np1/6B1/2PP2b1/3BPN2/PP3PPP/RN1Q1RK1 w - - 4 8h2h3;2kr3r/4q1p1/p1n2p2/7P/1PPp2P1/PBn2p2/2Q2P2/2R2K2 w - - 0 32c2f5:2kr4/ppp3pp/2P1p1q1/4np2/1P6/P3rB1P/2Q3P1/R4RK1 b - - 0 21e3f3;r4rk1/p2bbppp/1p6/2p1pP2/2B1N3/4B3/P1P3PP/3R1RK1 b - - 1 19f8d8;2kr1b1r/ppp3pp/5pb1/3q4/3n2P1/4B2P/PPPQBP2/2KR3R w - - 0 14d2d497k/p3b1pp/2p2n2/2Pp4/1P1Pp3/2N1B2P/5PP1/n2B2K1 w - - 3 26g1f1Dr1bq1rk1/ppp2ppp/","encoding":"utf8","key":"data/test/behavioral_cloning_data.bag","truncatedAt":512}],"sampleObjects":["data/test/action_value_data.bag","data/test/behavioral_cloning_data.bag","data/test/state_value_data.bag"],"sourceObjectKey":"data/test/action_value_data.bag","starterCell":"from google.cloud import storage\nclient = storage.Client.create_anonymous_client()\nbucket = client.bucket('searchless_chess')\nblob = bucket.blob('data/test/action_value_data.bag')\nwith blob.open(\"rb\") as fh:\n    print(fh.read(64))","starterSamplePath":"test/action_value_data.bag","totalObjects":"2 top-level prefixes"},"summary":"ChessBench is a benchmark of 10 million chess games with legal moves and value annotations from the Stockfish 16 engine, about 15 billion data points, released with the NeurIPS…","title":"ChessBench: 10 Million Chess Games with Stockfish 16 Action-Value, State-Value and Best-Move Annotations (Google DeepMind)","updatedAt":"2026-11-21T21:57:08.792Z","visibility":"public"}}