PBatch23888/chessq-chesscam

ChessQ ChessCam

ChessQ Lite fine-tuned on footage from ChessCam games and exported to LiteRT for the browser. It is the board-recognition model used by the ChessCam web app.

Built with ChessQueries (paper).

Files

• chessq-chesscam-fp16.tflite: ViT-S ChessQueries model, FP16 weights (64.3 MB). It runs with LiteRT.js on WebGPU or on the CPU (WebAssembly).
• chessq-chesscam-fp32.safetensors: the same model as PyTorch weights in FP32 (126 MB), in the ChessQueries checkpoint format.

Quick start (PyTorch)

With Python 3.12:

python -m pip install "git+https://github.com/JSeytre/chessqueries.git@f5aaaf38212cc7839b03ccf383455053905b3f7e"
chessqueries-predict photo.jpg --checkpoint chessq-chesscam-fp32.safetensors

Input and output

• Input: uint8 (1, 644, 644, 4), RGBA pixels (for example, canvas ImageData.data as is). The image is resized to 644 × 644 without keeping its aspect ratio. Normalisation happens inside the model.
• Outputs (their order is up to the runtime, so tell them apart by shape):
  • float32 (1, 64, 13): softmax probabilities for each square, with squares in FEN order (a8, b8, …, h1). The classes are . P N B R Q K p n b r q k, where . is an empty square.
  • float32 (1, 64, 2): where each square's query attends in the input image, as (x, y) in [0, 1], squares in the same order. It is the attention-weighted centre of the image patches in the last decoder layer's cross-attention, so it is pulled towards the board centre: use it for directions (for example, which way rank 8 lies), not exact square positions.

Training

The model starts from the released ChessQ Lite V4 weights and was trained for 4 epochs on the ChessReD and ChessCog training splits plus a ChessCam dataset of 15,590 frames from recorded games. The ChessCam labels were made by converting the app's piece boxes and board grids to FENs.

Comparison with ChessQ Lite

Test-set results for this model and ChessQ Lite, both run in PyTorch with FP32 arithmetic at 644 × 644.

Test set	Boards	Metric	ChessQ Lite	ChessQ ChessCam
ChessCam	2,275	Exact board	14.4%	38.8%
		At most 1 square wrong	33.1%	72.3%
		Squares correct	94.86%	98.25%
ChessReD	2,129	Exact board	98.6%	98.6%
		At most 1 square wrong	99.8%	99.9%
		Squares correct	99.97%	99.98%
ChessCog	342	Exact board	97.4%	97.4%
		At most 1 square wrong	100.0%	100.0%
		Squares correct	99.96%	99.96%

The ChessCam test videos are not in the training set. The test set only keeps frames where ChessQ Lite agrees with the labels on at least 48 of 64 squares, which favours ChessQ Lite and may not represent every kind of ChessCam footage. The .tflite file uses FP16 weights, so its results may differ slightly.

License

These weights are a modified version of ChessQ Lite and are licensed under the PolyForm Noncommercial License 1.0.0. They may be used, modified and redistributed for permitted noncommercial purposes; commercial use requires separate permission from the ChessQueries authors. Third-party components keep their upstream terms; see the notices.

Required Notice: Copyright (c) 2026 Joël Seytre.