Chaturanga Mini

Chaturanga is a lightweight SLM designed for playing chess using natural language move generation. It is based on the Supra family (https://huggingface.co/collections/SupraLabs/supra2) which allows it to generate (mostly)accurate and legal moves whilst keeping memory footprint to a minimum. It's small size (25M-100M parameters) allows it to run comfortably on laptop grade GPUs and CPUs.

Chaturanga Pro is a 100M variant based on the Supra-100M-Base which has been finetuned on 600,000 rows from the chessllm dataset.

The Pro variant is once again doubles the size of the Mini variant (50M -> 100M), which allows it to generate more comprehensive and non-illogical moves whilst keeping inference time relatively low. Despite its small size, it can still generate fairly precise and legal moves in SAN.

The Chaturanga model comes in the following sizes: Nano, Neo(Nano+), Mini, Pro and Ultra(Pro+)

Estimated parameters: ~100M

Architecture: Llama

Intended use: as an experimental chess-playing SLM

Training data

Source: chessllm dataset (https://huggingface.co/datasets/mlabonne/chessllm)
Rows: ~600,000 rows templated in SAN (Standard Algebraic Notation)
Training: trained for 7,000 steps on an RTX 3050 (4GB VRAM)

Usage

The model and the chess GUI can be found in the chaturanga_gui.py file included in the uploads.

Limitations

The model only has 100M parameters, so hallucinations and illegal moves may occur.
Model for experimental use only; users should employ it as such under license.