Downloads · 30 days
6
18% of all-time downloads
chess-transformer-200m-maxelo
avewright/chess-transformer-200m-maxelo
chess-transformer-200m-maxelo is a other model from avewright. Use it for the other task on the model card, and read the license before you ship it in a product. It is set up for pytorch. The card lists the license as mit.
Finetune of avewright/chess-transformer-200m-compact-soft for high Elo without MCTS: next-move prediction via legal-mask policy argmax.
All-time downloads
34
Public
Repo size
1.6 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt1.6 GB · 100%
At a glance
- Task
- other
- Library
- pytorch
- License
- mit
- Access
- Public
- Created
- Jul 11, 2026
- Updated
- Jul 11, 2026
- SHA
- 1c0e1b70
Model card
From the Hugging Face model README
Chess Transformer 200M — MaxElo (policy-only)
Finetune of avewright/chess-transformer-200m-compact-soft
for high Elo without MCTS: next-move prediction via legal-mask policy argmax.
Model card
| Architecture | ChessTransformer ~204M (fused board encoder, 16L × 1024d × 16H) |
| Vocab | compact (MOVE_VOCAB_VERSION=compact) |
| Heads | Spatial policy + 3-class WDL value |
| Inference | Policy argmax only (no search) |
| Base | chess-transformer-200m-compact-soft |
| Train run | exp189 (soft MultiPV + deep soft mix + hard depth≥15) |
Checkpoints in this repo
| File | Meaning |
|---|---|
best_model.pt | Best blended soft holdout top-1 during deep-mix resume |
latest_model.pt | Shutdown weights at step 2851 |
config.json | Architecture + training metadata |
PROGRESS.md | Full session write-up |
elo_eval.json | Raw Elo ladder results |
Lean checkpoints contain model_state_dict + metadata (no optimizer).
Training data mix
- Shallow soft — MultiPV soft targets from
avewright/exp186-sf-multipv-2m - Deep soft — phase-balanced SF18 MultiPV from
avewright/exp190-phase-deep-soft(~40% of soft steps) - Hard ballast — HF Stockfish-labeled stream,
min_depth ≥ 15
Augmentation: horizontal flip on soft batches (hflip_p=0.5).
Elo (pure policy)
Evaluated with elo_eval_latest.py vs Stockfish 18 UCI_LimitStrength
(50ms/move, opening book + Syzygy, 8 openings × both colors):
| Opponent Elo | Score | W–D–L |
|---|---|---|
| 1500 | 0.625 | 7–6–3 |
| 1800 | 0.438 | 4–6–6 |
Estimated Elo ≈ 1700 (bracket 1500–1800; small sample, noisy).
Quick load
import os, torch
os.environ["MOVE_VOCAB_VERSION"] = "compact"
from play import load_model # or elo_eval_latest.load_eval_model
model = load_model("best_model.pt", device="cuda")
model.eval()
Or play in the local GUI:
export MOVE_VOCAB_VERSION=compact
python play_factory_gui.py --checkpoint best_model.pt
Known limits
- Plateaued on the available soft mix; more deep phase-balanced data should help
- Weaker as Black in the Elo sample
- Castling indices use Chess960-style UCI in the vocab; GUI converts to standard UCI for chess.js
Citation / code
Training code: avewright/transform —
experiments/exp189_200m_maxelo_policy.py, docs/PROGRESS_2026-07-10.md.
- Task
- other
- Library
- pytorch
- License
- mit
- Languages
- en
- Created
- Jul 11, 2026
- Updated
- Jul 11, 2026