Python使用OpenAI调用Llama模型的通用教程(Llama2/Llama3/Llama3.1)

pip install openai

from openai import OpenAI

client = OpenAI(
    base_url="http://你的IP:8888/v1",
    api_key="你的部署密钥"
)

from openai import OpenAI

client = OpenAI(
    base_url="http://你的IP:8888/v1",
    api_key="你的部署密钥"
)

def llama_chat(question):
    response = client.chat.completions.create(
        model="Llama3.1-8B-Instruct",
        messages=[
            {"role":"system","content":"你是专业助手，回答准确、简洁、严格遵守用户要求"},
            {"role":"user","content": question}
        ],
        max_tokens=8192,
        temperature=0.1,
        top_p=0.3,
        frequency_penalty=0.05,
        presence_penalty=0.0,
        stream=False,
        extra_body={
            "top_k": 30,
            "repetition_penalty": 1.05
        }
    )
    return response.choices[0].message.content

if __name__ == "__main__":
    print(llama_chat("Python列表嵌套字典如何转为JSON字符串？"))

from openai import OpenAI

client = OpenAI(
    base_url="http://你的IP:8888/v1",
    api_key="你的部署密钥"
)

def llama_stream_chat(question):
    stream = client.chat.completions.create(
        model="Llama3.1-8B-Instruct",
        messages=[
            {"role":"system","content":"严格按照用户要求输出，无多余解释"},
            {"role":"user","content": question}
        ],
        max_tokens=8192,
        temperature=0.1,
        top_p=0.3,
        stream=True,
        extra_body={
            "top_k": 30,
            "repetition_penalty": 1.05
        }
    )

    full_text = ""
    for chunk in stream:
        if chunk.choices and chunk.choices[0].delta.content:
            text = chunk.choices[0].delta.content
            full_text += text
            print(text, end="", flush=True)
    return full_text

if __name__ == "__main__":
    llama_stream_chat("详细讲解大模型vLLM部署优势")

temperature=0.1,
top_p=0.3,
max_tokens=8192,
frequency_penalty=0.05,
stream=False,
extra_body={
    "top_k":30,
    "repetition_penalty":1.05
}

temperature=0.7,
top_p=0.8,
max_tokens=8192,
stream=True,
extra_body={
    "top_k":40,
    "repetition_penalty":1.03
}