Instructions to use GL3MON/Qwen3.5-2B-chess-finetuned with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use GL3MON/Qwen3.5-2B-chess-finetuned with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="GL3MON/Qwen3.5-2B-chess-finetuned")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("GL3MON/Qwen3.5-2B-chess-finetuned")
model = AutoModelForCausalLM.from_pretrained("GL3MON/Qwen3.5-2B-chess-finetuned", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use GL3MON/Qwen3.5-2B-chess-finetuned with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "GL3MON/Qwen3.5-2B-chess-finetuned"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "GL3MON/Qwen3.5-2B-chess-finetuned",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/GL3MON/Qwen3.5-2B-chess-finetuned

SGLang

How to use GL3MON/Qwen3.5-2B-chess-finetuned with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "GL3MON/Qwen3.5-2B-chess-finetuned" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "GL3MON/Qwen3.5-2B-chess-finetuned",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "GL3MON/Qwen3.5-2B-chess-finetuned" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "GL3MON/Qwen3.5-2B-chess-finetuned",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use GL3MON/Qwen3.5-2B-chess-finetuned with Docker Model Runner:
```
docker model run hf.co/GL3MON/Qwen3.5-2B-chess-finetuned
```

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Qwen-3.5-2B Chess Fine-tuned (SFT)

Overview

This is a fine-tuned version of Qwen-3.5-2B on chess move prediction using Supervised Fine-Tuning (SFT). The model was trained on the Thytu/ChessInstruct dataset.

Model Details

Base Model: Qwen/Qwen3.5-2B
Fine-tuning Method: Supervised Fine-Tuning (SFT)
Precision: bfloat16

Training Configuration

Learning Rate: 1e-5
Batch Size: 4 (per device)
Gradient Accumulation Steps: 1
Epochs: 3
Optimizer: AdamW with cosine learning rate scheduler
Warmup Steps: 100
Weight Decay: 0.01

Training Dynamics

Dataset

The model was trained on Thytu/ChessInstruct, a chess instruction dataset containing:

Chess positions with move annotations
Game sequences with context
Various chess task categories (move prediction, advantage evaluation, etc.)

Task Categories

Category	Description
`FIND_NEXT_BEST_MOVE`	Predict the next best move
`FIND_ADVANTAGED_PLAYER`	Identify who has advantage
`FIND_FINAL_SCORE`	Predict game result
`MLM_ON_MOVES`	Fill in missing moves
`FIND_LAST_MOVE`	Find the final move
`SORT_FENS`	Sort positions by game order

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_name = "GL3MON/Qwen3.5-2B-chess-finetuned"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.bfloat16,
    device_map="auto"
)

# Example: Predict chess moves
messages = [
    {"role": "system", "content": "You are a chess assistant. Provide the best move in SAN format."},
    {"role": "user", "content": "What is the best move for white in the position: rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1?"}
]

input_ids = tokenizer.apply_chat_template(messages, return_tensors="pt").to(model.device)
outputs = model.generate(input_ids, max_new_tokens=50, temperature=0.7, do_sample=True)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)

Evaluation

The model was evaluated on a validation split of the ChessInstruct dataset. The training achieved good convergence with the loss decreasing steadily over 3 epochs.

License

MIT License

Citation

@software{ChessLM2026,
  title = {ChessLM - Qwen-3.5-2B Chess Assistant},
  author = {GL3MON},
  year = {2026},
  url = {https://huggingface.co/GL3MON/Qwen3.5-2B-chess-finetuned}
}