from transformers import AutoModelForCausalLM, AutoTokenizer,set_seed
set_seed(42)

model_name = "twnlp/ChineseErrorCorrector3-4B"

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

prompt = "你是一个文本纠错专家，纠正输入句子中的语法错误，并输出正确的句子，输入句子为："
text_input = "对待每一项工作都要一丝不够。"
messages = [
    {"role": "user", "content": prompt + text_input}
]
text = tokenizer.apply_chat_template(
        messages,
        tokenize=False,
        add_generation_prompt=True,
        enable_thinking=False # Switches between thinking and non-thinking modes. Default is True.
    )
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=512
)
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)

pip install transformers
pip install vllm==0.8.5

from transformers import AutoTokenizer
from vllm import LLM, SamplingParams

# Initialize the tokenizer
tokenizer = AutoTokenizer.from_pretrained("twnlp/ChineseErrorCorrector3-4B")

# Pass the default decoding hyperparameters of twnlp/ChineseErrorCorrector3-4B
# max_tokens is for the maximum length for generation.
sampling_params = SamplingParams(seed=42,max_tokens=512)

# Input the model name or path. Can be GPTQ or AWQ models.
llm = LLM(model="twnlp/ChineseErrorCorrector3-4B")

# Prepare your prompts
text_input = "对待每一项工作都要一丝不够。"
messages = [
    {"role": "user", "content": "你是一个文本纠错专家，纠正输入句子中的语法错误，并输出正确的句子，输入句子为："+text_input}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=False
)

# generate outputs
outputs = llm.generate([text], sampling_params)

# Print the outputs.
for output in outputs:
    prompt = output.prompt
    generated_text = output.outputs[0].text
    print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}") 

git clone https://github.com/TW-NLP/ChineseErrorCorrector
cd ChineseErrorCorrector

conda create -n zh_correct -y python=3.10
conda activate zh_correct
pip install -r requirements.txt
# If you are in mainland China, you can set the mirror as follows:
pip install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple/ --trusted-host=mirrors.aliyun.com

CUDA_VISIBLE_DEVICES=0 nohup vllm serve twnlp/ChineseErrorCorrector3-4B \
    --port 8000 \
    --max-model-len 1024 \
    --gpu-memory-utilization 0.9 \
    --seed 42 \
    >chinese_corrector.log 2>&1 &

curl http://localhost:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "twnlp/ChineseErrorCorrector3-4B",
    "messages": [
      {
        "role": "user",
        "content": "你是一个文本纠错专家，纠正输入句子中的语法错误，并输出正确的句子，输入句子为：对待每一项工作都要一丝不够。"
      }
    ],
    "max_tokens": 1024,
    "temperature": 0,
    "seed": 42
  }'

# 修改config.py
#（1）根据不同的模型，修改的DEFAULT_CKPT_PATH，默认为twnlp/ChineseErrorCorrector3-4B(将模型下载，放在ChineseErrorCorrector/pre_model/twnlp/ChineseErrorCorrector3-4B)
#（2）将TextCorrectConfig的USE_VLLM = True

#批量预测
python main.py
#输出：
'''
[{'source': '对待每一项工作都要一丝不够。', 'target': '对待每一项工作都要一丝不苟。', 'errors': [('够', '苟', 12)]}, {'source': '大约半个小时左右', 'target': '大约半个小时', 'errors': [('左右', '', 6)]}]
'''

git clone https://github.com/TW-NLP/ChineseErrorCorrector
cd ChineseErrorCorrector

conda create -n zh_correct -y python=3.10
conda activate zh_correct
pip install -r requirements.txt
# If you are in mainland China, you can set the mirror as follows:
pip install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple/ --trusted-host=mirrors.aliyun.com

# 修改config.py
#（1）根据不同的模型，修改的DEFAULT_CKPT_PATH，默认为twnlp/ChineseErrorCorrector3-4B
#（2）将TextCorrectConfig的USE_VLLM = False

#批量预测
python main.py

from modelscope import AutoModelForCausalLM, AutoTokenizer

model_name = "tiannlp/twnlp/ChineseErrorCorrector3-4B"

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

prompt = "你是一个文本纠错专家，纠正输入句子中的语法错误，并输出正确的句子，输入句子为："
text_input = "对待每一项工作都要一丝不够。"
messages = [
    {"role": "user", "content": prompt + text_input}
]
text = tokenizer.apply_chat_template(
        messages,
        tokenize=False,
        add_generation_prompt=True,
        enable_thinking=False # Switches between thinking and non-thinking modes. Default is True.
    )
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=512
)
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)

@misc{tian2025chineseerrorcorrector34bstateoftheartchinesespelling,
      title={ChineseErrorCorrector3-4B: State-of-the-Art Chinese Spelling and Grammar Corrector}, 
      author={Wei Tian and YuhaoZhou},
      year={2025},
      eprint={2511.17562},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2511.17562}, 
}