Instructions to use starriver030515/Qwen2.5-Math-7B-32k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use starriver030515/Qwen2.5-Math-7B-32k with Transformers:

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

pipe = pipeline("text-generation", model="starriver030515/Qwen2.5-Math-7B-32k")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("starriver030515/Qwen2.5-Math-7B-32k")
model = AutoModelForCausalLM.from_pretrained("starriver030515/Qwen2.5-Math-7B-32k")
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 starriver030515/Qwen2.5-Math-7B-32k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "starriver030515/Qwen2.5-Math-7B-32k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "starriver030515/Qwen2.5-Math-7B-32k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/starriver030515/Qwen2.5-Math-7B-32k

SGLang

How to use starriver030515/Qwen2.5-Math-7B-32k 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 "starriver030515/Qwen2.5-Math-7B-32k" \
    --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": "starriver030515/Qwen2.5-Math-7B-32k",
		"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 "starriver030515/Qwen2.5-Math-7B-32k" \
        --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": "starriver030515/Qwen2.5-Math-7B-32k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use starriver030515/Qwen2.5-Math-7B-32k with Docker Model Runner:
```
docker model run hf.co/starriver030515/Qwen2.5-Math-7B-32k
```

The base Qwen2.5-Math-7B model used by HAPO. We extend the context window to 32k.

Citation

If you find our model, data, or evaluation code useful, please kindly cite our paper:

@misc{liu2025uniformheterogeneoustailoringpolicy,
      title={From Uniform to Heterogeneous: Tailoring Policy Optimization to Every Token's Nature}, 
      author={Zheng Liu and Mengjie Liu and Siwei Wen and Mengzhang Cai and Bin Cui and Conghui He and Wentao Zhang},
      year={2025},
      eprint={2509.16591},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2509.16591}, 
}

Downloads last month: 2

Safetensors

Model size

8B params

Tensor type

BF16

Collection including starriver030515/Qwen2.5-Math-7B-32k

HAPO

Collection

4 items • Updated Sep 23, 2025

Paper for starriver030515/Qwen2.5-Math-7B-32k

From Uniform to Heterogeneous: Tailoring Policy Optimization to Every Token's Nature

Paper • 2509.16591 • Published Sep 20, 2025 • 2