---
title: README
emoji: 🐢
colorFrom: indigo
colorTo: yellow
sdk: gradio
pinned: false
---

# Leela Chess Zero

All models here are "Chessformers" — encoder-only transformers with the Geometric Attention Bias described in the paper [Chessformer: A Unified Architecture for Chess Modeling](https://openreview.net/pdf?id=2ltBRzEHyd).

They are expressed in the internal Lc0 format built on Protobuf (hence `.pb.gz`), which is compatible with the training code and the engine. They can be converted to ONNX through the `leela2onnx` command through the Lc0 binary, which we recommend for mechanistic interpretability research.

**BT4** is our largest model at 190M and corresponds to Leela-CF. BT4 and BT3 were trained with SFT on a prior RL run, and the T1 models were distilled from a larger transformer. The models range in skill from expert-level (T1-10L-256D) to grandmaster-plus-level (BT4) at Blitz time controls with greedy sampling of policy prediction.

See also the [website](lczero.org), [Discord server](lczero.org/chat), and [a longest model list](https://lczero.org/dev/wiki/best-nets-for-lc0/) of the Leela Chess Zero project.