ChessCheatDetector.com
Newv3 model, trained on real grandmaster games →Chess Cheat Detector
Analyze a chess game for signs of engine assistance. Paste a Chess.com or Lichess link, or upload a PGN.
We run every move through Stockfish, then a custom machine-learning model looks for play that is unusually engine-like.
Engine assistance likelihood
- Top engine choice
- 88%
- Engine top 3
- 97%
- Top choice when critical
- 93%
- Average centipawn loss
- 7
- Mistakes
- 0%
- Moves scored
- 32
This is illustrative data only.
Custom AI-powered cheat detection
No LLMs. The machine-learning model was built from scratch for one job: detecting cheating in chess.
New in v3: it's trained on thousands of real grandmaster games, so honest strong players are 16× less likely to be mistaken for an engine.
- Model
- Custom ML v3, built from scratch
- LLMs
- None
- Inputs
- Per-move engine and positional features
- Rating-aware
- Judges moves against human play at your rating
- Target
- Engine assistance, not move quality
- Status
- Live
How chess cheat detection works
- 01
Parse
The game is read into positions, players, ratings and results, from a PGN or a Chess.com or Lichess link.
- 02
Engine
Stockfish scores every position on our server.
- 03
Features
Centipawn loss, agreement with the engine, details of each position, and how likely a player at that rating is to play each move.
- 04
Detector
Our custom model turns those features and both players' ratings into a suspicion profile.