使用 Zep 开源长久记忆服务为 AI 助手构建个性化体验

最新推荐文章于 2025-12-07 19:40:17 发布

原创最新推荐文章于 2025-12-07 19:40:17 发布 · 395 阅读

4 ·

CC 4.0 BY-SA版权

文章标签：

#人工智能 #python

技术背景介绍

随着人工智能助手的广泛应用，如何让 AI 能够记住用户的历史对话，提供更个性化的互动体验，已经成为一个重要的技术挑战。Zep 是一种专为 AI 助手应用设计的长久记忆服务，可以帮助 AI 助手在对话中抓取、理解和提取历史数据，从而增强其个性化服务能力。

核心原理解析

Zep 通过提供一个可检索和自动总结的对话历史存储，帮助减少 AI 的幻觉（hallucinations）、降低延迟和成本。Zep 的长久记忆功能使得即便是较为久远的对话内容，也能被回忆和检索。

代码实现演示

下文将演示如何使用 Zep 来为你的聊天机器人提供记忆功能。

from uuid import uuid4
from langchain.agents import AgentType, initialize_agent
from langchain.memory import ZepMemory
from langchain_community.retrievers import ZepRetriever
from langchain_community.utilities import WikipediaAPIWrapper
from langchain_core.messages import AIMessage, HumanMessage
from langchain_core.tools import Tool
from langchain_openai import OpenAI

# Zep 服务器的 URL
ZEP_API_URL = "http://localhost:8000"

session_id = str(uuid4())  # 用户的唯一标识

# 获取您的 OpenAI API 密钥
import getpass

openai_key = getpass.getpass()

# 获取 Zep API 密钥(可选)
zep_api_key = getpass.getpass()

# 初始化 Zep 聊天历史类和代理
search = WikipediaAPIWrapper()
tools = [
    Tool(
        name="Search",
        func=search.run,
        description=(
            "当你需要在线搜索答案时非常有用。你应该提出"
            "针对性的问题"
        ),
    ),
]

# 配置 Zep 聊天历史
memory = ZepMemory(
    session_id=session_id,
    url=ZEP_API_URL,
    api_key=zep_api_key,
    memory_key="chat_history",
)

# 初始化代理
llm = OpenAI(temperature=0, openai_api_key=openai_key)
agent_chain = initialize_agent(
    tools,
    llm,
    agent=AgentType.CONVERSATIONAL_REACT_DESCRIPTION,
    verbose=True,
    memory=memory,
)

# 预加载对话历史数据
test_history = [
    {"role": "human", "content": "Who was Octavia Butler?"},
    {
        "role": "ai",
        "content": (
            "Octavia Estelle Butler (June 22, 1947 – February 24, 2006) was an American"
            " science fiction author."
        ),
    },
    {"role": "human", "content": "Which books of hers were made into movies?"},
    {
        "role": "ai",
        "content": (
            "The most well-known adaptation of Octavia Butler's work is the FX series"
            " Kindred, based on her novel of the same name."
        ),
    },
    # 略去部分数据...
]

for msg in test_history:
    memory.chat_memory.add_message(
        (
            HumanMessage(content=msg["content"])
            if msg["role"] == "human"
            else AIMessage(content=msg["content"])
        ),
        metadata=msg.get("metadata", {}),
    )

# 运行代理，将输入和响应自动添加到 Zep 记忆中。
agent_chain.run(
    input="What is the book's relevance to the challenges facing contemporary society?",
)

# 检查 Zep 记忆
def print_messages(messages):
    for m in messages:
        print(m.type, ":\n", m.dict())

print(memory.chat_memory.zep_summary)
print("\n")
print_messages(memory.chat_memory.messages)