You can deploy this model and start earning money today!

Pradheep1647/
kibitzer-sft-elo-1079-step-012000

Name: kibitzer-sft-elo-1079-step-012000
Availability: InStock

Views

No views yet

Pradheep1647/kibitzer-sft-elo-1079-step-012000

Kibitzer checkpoint for a chess policy/value model trained on game sequences. Each board is encoded as 64 square tokens plus auxiliary state, summarized by a square-level encoder, then processed by a causal transformer over the game timeline.

Status

checkpoint: step_012000.pt
step: 012000
elo_rating: 1079 estimated vs stockfish-elo-1320 (-240.8 Elo diff)
github_repo: https://github.com/Mantissagithub/kibitzer
eval_opponent: stockfish-elo-1320
elo_diff: -240.8
eval_games: 20
eval_score: 4.00
elo_error: 274.1

Architecture

Input: 64 chess-square piece tokens plus 7 auxiliary scalars covering side to move, castling rights, en-passant file, and halfmove clock.
Position encoder: bidirectional square-level transformer that compresses each board into one timeline token.
Timeline trunk: causal transformer with d_model=384, sequence length 256, 12 layers, and 8 attention heads.
Transformer blocks: pre-norm RMSNorm, RoPE on causal Q/K attention, PyTorch scaled dot-product attention, and SwiGLU feed-forward layers.
Outputs: policy head over 4,672 AlphaZero-style moves (64 * 73) plus a tanh value head for bounded outcome prediction.

Files

step_012000.pt: PyTorch checkpoint.
training_metadata.yaml: training config and checkpoint metrics.
post_eval.yaml: uploaded after local Stockfish/cutechess eval.

Usage

bash

git clone https://github.com/Mantissagithub/kibitzer.git
cd kibitzer
uv run python scripts/uci.py --checkpoint <path-to>/step_012000.pt

This checkpoint is one artifact in a sequence. Repos with elo-pending in the name have not been evaluated yet; rated repos are renamed after scripts/eval_and_rename_hf.py --from-hf completes.