I hope you can choose the QA of the file itself for uploading

[401557122]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）
• Please do not modify this template :) and fill in all the required fields.

1. Is this request related to a challenge you're experiencing? Tell me about your story.

In order to avoid using a large model to generate QA pairs, which is too slow, I modified the following code and standardized the file header. However, I did not use LLMS during QA preview. Why did I still use a large model to generate QA during encoding and the final result
def generate_qa_document(cls, tenant_id: str, query, document_language: str):
prompt = GENERATOR_QA_PROMPT.format(language=document_language)
#首先使用正则表达式提取原始问答
import re
match = re.search(r'questions:\s*(.?)\s;\sanswers:\s(.*)', query, re.DOTALL)
if match:
one_question = match.group(1).strip()
one_answer = match.group(2).strip()
answer = 'Q1:'+one_question+'\n'+'A1:'+one_answer
else:
model_manager = ModelManager()
model_instance = model_manager.get_default_model_instance(
tenant_id=tenant_id,
model_type=ModelType.LLM,
)

        prompt_messages = [
            SystemPromptMessage(content=prompt),
            UserPromptMessage(content=query)
        ]

        response = model_instance.invoke_llm(
            prompt_messages=prompt_messages,
            model_parameters={
                'temperature': 0.01,
                "max_tokens": 2000
            },
            stream=False
        )

        answer = response.message.content
    return answer.strip()

2. Additional context or comments

No response

3. Can you help us with this feature?

• I am interested in contributing to this feature.

[crazywoola]

Duplicated #6904