Streamlit集成LangChain ConversationBufferMemory无法记录对话历史问题

核心问题 #

1. 内存对象未持久化：chain在全局作用域初始化，每次Streamlit页面刷新（比如输入内容后）都会重新调用load_chain创建新的ConversationBufferMemory，之前的对话历史被清空。
2. 手动传入history参数冲突：调用chain.run(input=user_input, history=st.session_state["history"])时，手动传入的history会覆盖LangChain内存组件自动管理的对话历史，导致ConversationBufferMemory未生效。
3. Prompt配置不匹配：使用ConversationChain时，自定义Prompt的MessagesPlaceholder变量名未与内存的memory_key对齐，导致历史对话无法正确注入。

修复后的完整代码 #

import streamlit as st
from streamlit_chat import message

from langchain.chains import ConversationChain
from langchain.chat_models import AzureChatOpenAI
from langchain.memory import ConversationBufferMemory
from langchain.prompts import (
    ChatPromptTemplate, 
    MessagesPlaceholder, 
    SystemMessagePromptTemplate, 
    HumanMessagePromptTemplate
)

# 初始化Session State，确保chain和内存持久化
if "chain" not in st.session_state:
    prompt = ChatPromptTemplate.from_messages([
        SystemMessagePromptTemplate.from_template("The following is a friendly conversation between a human and an AI. The AI is talkative and provides lots of specific details from its context. If the AI does not know the answer to a question, it truthfully says it does not know."),
        MessagesPlaceholder(variable_name="history"),  # 和memory的memory_key对应
        HumanMessagePromptTemplate.from_template("{input}")
    ])
    
    llm = AzureChatOpenAI(
                    deployment_name='gpt-35-turbo',
                    model_name='gpt-35-turbo',
                    temperature=0,
                    openai_api_key='.....',
                    openai_api_base='.....',
                    openai_api_version="2023-05-15", 
                    openai_api_type="azure"
                    )
    # 指定memory_key为"history"，和Prompt中的变量名一致
    memory = ConversationBufferMemory(memory_key="history", return_messages=True)
    st.session_state["chain"] = ConversationChain(
        llm=llm,
        verbose=True,
        prompt=prompt,
        memory=memory
    )

# Streamlit页面配置
st.set_page_config(page_title="LangChain Demo", page_icon=":robot:")
st.header("LangChain Demo")

if "generated" not in st.session_state:
    st.session_state["generated"] = []

if "past" not in st.session_state:
    st.session_state["past"] = []

def get_text():
    input_text = st.text_input("You: ", "Hello, how are you?", key="input")
    return input_text

user_input = get_text()

if user_input:
    # 直接调用chain.run，只传input参数，内存由LangChain自动管理
    output = st.session_state["chain"].run(input=user_input)
    st.session_state.past.append(user_input)
    st.session_state.generated.append(output)

if st.session_state["generated"]:
    for i in range(len(st.session_state["generated"]) - 1, -1, -1):
        message(st.session_state["generated"][i], key=str(i))
        message(st.session_state["past"][i], is_user=True, key=str(i) + "_user")

关键修复点说明 #

• 持久化chain和内存：把chain的初始化放到st.session_state中，避免页面刷新时重新创建内存对象，保证对话历史留存。
• 移除手动history参数：调用chain.run时只传入input，LangChain会自动从ConversationBufferMemory中读取历史对话并填充到Prompt的history占位符。
• 对齐memory_key和Prompt变量名：创建ConversationBufferMemory时指定memory_key="history"，和Prompt中MessagesPlaceholder(variable_name="history")保持一致，确保内存中的历史能正确传入Prompt。

内容的提问来源于stack exchange，提问作者illuminato