Instructions to use BitStarWalkin/SuperCorrect-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BitStarWalkin/SuperCorrect-7B with Transformers:

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

pipe = pipeline("text-generation", model="BitStarWalkin/SuperCorrect-7B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("BitStarWalkin/SuperCorrect-7B")
model = AutoModelForCausalLM.from_pretrained("BitStarWalkin/SuperCorrect-7B")
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 BitStarWalkin/SuperCorrect-7B with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/BitStarWalkin/SuperCorrect-7B

SGLang

How to use BitStarWalkin/SuperCorrect-7B 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 "BitStarWalkin/SuperCorrect-7B" \
    --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": "BitStarWalkin/SuperCorrect-7B",
		"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 "BitStarWalkin/SuperCorrect-7B" \
        --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": "BitStarWalkin/SuperCorrect-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use BitStarWalkin/SuperCorrect-7B with Docker Model Runner:
```
docker model run hf.co/BitStarWalkin/SuperCorrect-7B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

SuperCorrect: Supervising and Correcting Language Models with Error-Driven Insights

SuperCorrect: Supervising and Correcting Language Models with Error-Driven Insights Ling Yang*, Zhaochen Yu*, Tianjun Zhang, Minkai Xu, Joseph E. Gonzalez,Bin Cui, Shuicheng Yan

Peking University, Skywork AI, UC Berkeley, Stanford University

Introduction

This repo provides the official implementation of SuperCorrect a novel two-stage fine-tuning method for improving both reasoning accuracy and self-correction ability for LLMs.

Notably, our SupperCorrect-7B model significantly surpasses powerful DeepSeekMath-7B by 7.8%/5.3% and Qwen2.5-Math-7B by 15.1%/6.3% on MATH/GSM8K benchmarks, achieving new SOTA performance among all 7B models.

🚨 Unlike other LLMs, we incorporate LLMs with our pre-defined hierarchical thought template ([Buffer of Thought (BoT)](https://github.com/YangLing0818/buffer-of-thought-llm)) to conduct more deliberate reasoning than conventional CoT. It should be noted that our evaluation methods relies on pure mathematical reasoning abilities of LLMs, instead of leverage other programming methods such as PoT and ToRA.

Examples

🚨 For more concise and clear presentation, we omit some XML tags.

Model details

You can check our Github repo for more details.

Quick Start

Requirements

Since our current model is based on Qwen2.5-Math series, transformers>=4.37.0 is needed for Qwen2.5-Math models. The latest version is recommended.

🚨 This is a must because `transformers` integrated Qwen2 codes since `4.37.0`.

Inference

🤗 Hugging Face Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "BitStarWalkin/SuperCorrect-7B"
device = "cuda" 

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

prompt = "Find the distance between the foci of the ellipse \[9x^2 + \frac{y^2}{9} = 99.\]"
hierarchical_prompt = "Solve the following math problem in a step-by-step XML format, each step should be enclosed within tags like <Step1></Step1>. For each step enclosed within the tags, determine if this step is challenging and tricky, if so, add detailed explanation and analysis enclosed within <Key> </Key> in this step, as helpful annotations to help you thinking and remind yourself how to conduct reasoning correctly. After all the reasoning steps, summarize the common solution and reasoning steps to help you and your classmates who are not good at math generalize to similar problems within <Generalized></Generalized>. Finally present the final answer within <Answer> </Answer>."
# HT
messages = [
    {"role": "system", "content":hierarchical_prompt },
    {"role": "user", "content": prompt}
]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=1024
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(response)

Performance

We evaluate our SupperCorrect-7B on two widely used English math benchmarks GSM8K and MATH. All evaluations are tested with our evaluation method which is zero-shot hierarchical thought based prompting.

Citation

@article{yang2024supercorrect,
title={SuperCorrect: Supervising and Correcting Language Models with Error-Driven Insights}
  author={Yang, Ling and Yu, Zhaochen and Zhang, Tianjun and Xu, Minkai and Gonzalez, Joseph E and Cui, Bin and Yan, Shuicheng},
  journal={arXiv preprint arXiv:2410.09008},
  year={2024}
}
@article{yang2024buffer,
  title={Buffer of Thoughts: Thought-Augmented Reasoning with Large Language Models},
  author={Yang, Ling and Yu, Zhaochen and Zhang, Tianjun and Cao, Shiyi and Xu, Minkai and Zhang, Wentao and Gonzalez, Joseph E and Cui, Bin},
  journal={arXiv preprint arXiv:2406.04271},
  year={2024}
}

Acknowledgements

Our SuperCorrect is a two-stage fine-tuning model which based on several extraordinary open-source models like Qwen2.5-Math, DeepSeek-Math, Llama3-Series. Our evaluation method is based on the code base of outstanding works like Qwen2.5-Math and lm-evaluation-harness. We also want to express our gratitude for amazing works such as BoT which provides the idea of thought template.