Instructions to use zhoudoe23/ChessQween3-base-puzzled with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zhoudoe23/ChessQween3-base-puzzled with Transformers:

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

pipe = pipeline("text-generation", model="zhoudoe23/ChessQween3-base-puzzled")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("zhoudoe23/ChessQween3-base-puzzled")
model = AutoModelForCausalLM.from_pretrained("zhoudoe23/ChessQween3-base-puzzled", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use zhoudoe23/ChessQween3-base-puzzled with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zhoudoe23/ChessQween3-base-puzzled"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zhoudoe23/ChessQween3-base-puzzled",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/zhoudoe23/ChessQween3-base-puzzled

SGLang

How to use zhoudoe23/ChessQween3-base-puzzled 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 "zhoudoe23/ChessQween3-base-puzzled" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zhoudoe23/ChessQween3-base-puzzled",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "zhoudoe23/ChessQween3-base-puzzled" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zhoudoe23/ChessQween3-base-puzzled",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use zhoudoe23/ChessQween3-base-puzzled with Docker Model Runner:
```
docker model run hf.co/zhoudoe23/ChessQween3-base-puzzled
```

ChessQween3-base-puzzled

This model is a fine-tuned version of zhoudoe23/ChessQween3-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.8700

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 3e-05
train_batch_size: 128
eval_batch_size: 64
seed: 42
distributed_type: multi-GPU
num_devices: 2
gradient_accumulation_steps: 2
total_train_batch_size: 512
total_eval_batch_size: 128
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 1000
num_epochs: 1
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.1666	0.0233	1000	1.1704
1.0454	0.0466	2000	1.1195
1.1005	0.0700	3000	1.0880
1.0877	0.0933	4000	1.0663
1.0387	0.1166	5000	1.0488
1.1939	0.1399	6000	1.0355
1.1149	0.1633	7000	1.0199
1.0303	0.1866	8000	1.0077
1.0135	0.2099	9000	0.9969
1.0281	0.2332	10000	0.9883
0.9677	0.2566	11000	0.9776
1.0151	0.2799	12000	0.9704
0.9752	0.3032	13000	0.9628
0.9267	0.3265	14000	0.9540
1.0725	0.3499	15000	0.9486
1.0237	0.3732	16000	0.9456
0.9124	0.3965	17000	0.9355
0.8746	0.4198	18000	0.9291
1.0226	0.4431	19000	0.9235
0.8150	0.4665	20000	0.9190
0.9453	0.4898	21000	0.9152
0.8654	0.5131	22000	0.9089
1.0067	0.5364	23000	0.9057
0.8130	0.5598	24000	0.9016
0.9912	0.5831	25000	0.8974
0.8276	0.6064	26000	0.8950
0.9481	0.6297	27000	0.8912
0.9450	0.6531	28000	0.8884
0.8170	0.6764	29000	0.8853
0.9074	0.6997	30000	0.8833
0.9123	0.7230	31000	0.8817
0.8216	0.7464	32000	0.8783
0.8683	0.7697	33000	0.8765
0.8481	0.7930	34000	0.8748
0.8536	0.8163	35000	0.8735
0.7900	0.8397	36000	0.8729
0.8798	0.8630	37000	0.8718
0.8715	0.8863	38000	0.8712
0.8726	0.9096	39000	0.8705
0.8342	0.9329	40000	0.8702
0.8996	0.9563	41000	0.8700
0.8560	0.9796	42000	0.8700

Framework versions

Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 5.0.0
Tokenizers 0.22.2

Downloads last month: 156

Safetensors

Model size

0.1B params

Tensor type

F32

Model tree for zhoudoe23/ChessQween3-base-puzzled

Base model

zhoudoe23/ChessQween3-base

Finetuned

(1)

this model

zhoudoe23
/

ChessQween3-base-puzzled

ChessQween3-base-puzzled

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Model tree for zhoudoe23/ChessQween3-base-puzzled

Space using zhoudoe23/ChessQween3-base-puzzled 1