Instructions to use QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k with Transformers:

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

pipe = pipeline("text-generation", model="QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k")
model = AutoModelForCausalLM.from_pretrained("QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k")
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

vLLM

How to use QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k

SGLang

How to use QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k 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 "QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k" \
    --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": "QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k",
		"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 "QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k" \
        --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": "QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k with Docker Model Runner:
```
docker model run hf.co/QizhiPei/Qwen2.5-Math-7B-Instruct-RoPE-300k
```

Qwen2.5-Math-7B-RoPE-300k

This model is a variant of Qwen/Qwen2.5-Math-7B-Instruct, with its RoPE base frequency raised from 10k to 300k, enabling the context length to expand from 4k to 32k tokens.

Citation

If you find this model useful, please cite the original Qwen2.5-Math paper:

@article{yang2024qwen2,
  title={Qwen2. 5-math technical report: Toward mathematical expert model via self-improvement},
  author={Yang, An and Zhang, Beichen and Hui, Binyuan and Gao, Bofei and Yu, Bowen and Li, Chengpeng and Liu, Dayiheng and Tu, Jianhong and Zhou, Jingren and Lin, Junyang and others},
  journal={arXiv preprint arXiv:2409.12122},
  year={2024}
}