Instructions to use xiaodongguaAIGC/xdg-math-step with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use xiaodongguaAIGC/xdg-math-step with Transformers:

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

pipe = pipeline("text-generation", model="xiaodongguaAIGC/xdg-math-step")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("xiaodongguaAIGC/xdg-math-step")
model = AutoModelForCausalLM.from_pretrained("xiaodongguaAIGC/xdg-math-step")

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use xiaodongguaAIGC/xdg-math-step with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/xiaodongguaAIGC/xdg-math-step

SGLang

How to use xiaodongguaAIGC/xdg-math-step 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 "xiaodongguaAIGC/xdg-math-step" \
    --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": "xiaodongguaAIGC/xdg-math-step",
		"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 "xiaodongguaAIGC/xdg-math-step" \
        --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": "xiaodongguaAIGC/xdg-math-step",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use xiaodongguaAIGC/xdg-math-step with Docker Model Runner:
```
docker model run hf.co/xiaodongguaAIGC/xdg-math-step
```

小冬瓜AIGC：Step-Wise数学推理

Test：Colab

result

测试可以rejection sampling多次，以\boxed{}格式输出final asnwer

prompt = 'Tom has 12 apples. He gives 3 apples to each of his 4 friends. After that, he buys 10 more apples. How many apples does Tom have now?'
step_generation(prompt, 128)

result

<|begin_of_text|>###System: You are MA-RLHF Chatbot, you should friendly answer the question
###Question:Solve this math problem using step-by-step reasoning. Require that the output of each step ends with the " [SEP]
" token.
Tom has 12 apples. He gives 3 apples to each of his 4 friends. After that, he buys 10 more apples. How many apples does Tom have now?
###Answer: At first, Tom has 12 apples. [SEP]
He gives 3 apples to each of his 4 friends, so he gives him 4 * 3 = 12 apples. [SEP]
After that, Tom has 12 - 12 = 0 apples left. [SEP]
He buys 10 more apples, so he has 0 + 10 = 10 apples now. [SEP]
Tom has 10 apples now. [SEP]
Answer: 10 [SEP]
I agree. [SEP]
A possible answer.

# Answer

10 [SEP]
<|end_of_text|>

Downloads last month: 5

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for xiaodongguaAIGC/xdg-math-step

Base model

meta-llama/Llama-3.1-8B

Finetuned

(1786)

this model

Adapters

1 model

xiaodongguaAIGC
/

xdg-math-step

小冬瓜AIGC：Step-Wise数学推理

result

Model tree for xiaodongguaAIGC/xdg-math-step

Dataset used to train xiaodongguaAIGC/xdg-math-step