人工智能-LangGraph+ChatUI+DeepSeek API搭建本地智能助手

最新推荐文章于 2025-05-24 19:24:50 发布

原创最新推荐文章于 2025-05-24 19:24:50 发布 · 695 阅读

9 ·

CC 4.0 BY-SA版权

文章标签：

#人工智能

AI 专栏收录该内容

9 篇文章

订阅专栏

人工智能-LangGraph+ChatUI+DeepSeek API搭建本地智能助手

0 环境说明

环境项	环境说明
操作系统	Windows11 专业版
硬件信息	联想拯救者Y9000P
conda	Ananconda
Python版本	3.12
NodeJs	18.20.0

# 使用conda创建python环境
conda create -n langgraph python=3.12

# 激活环境
conda activate langgraph

# 下载依赖
pip install langchain-core langgraph -i https://pypi.tuna.tsinghua.edu.cn/simple

pip install "langgraph-cli[inmem]" --upgrade -i https://pypi.tuna.tsinghua.edu.cn/simple

1 LangGraph

下载一个LangGraph Template项目

git clone https://github.com/langchain-ai/new-langgraph-project.git

修改配置文件

cd new-langgraph-project
cp .env.example .env

因为我使用的是DeepSeek API，所以需要配置 DEEPSEEK_API_KEY（上一步创建的DeepSeek API Key）

# To separate your traces from other application
LANGSMITH_PROJECT=new-agent

# The following depend on your selected configuration

## LLM choice:
# ANTHROPIC_API_KEY=....
# FIREWORKS_API_KEY=...
# OPENAI_API_KEY=...

DEEPSEEK_API_KEY=your-deepseek-api-key

修改代码
src/agent/state.py

from __future__ import annotations

from typing_extensions import TypedDict

from dataclasses import dataclass
from langchain_core.messages import AnyMessage
from langgraph.graph.message import add_messages
from typing import Annotated


@dataclass
class State(TypedDict):
    messages: Annotated[list[AnyMessage], add_messages]

src/agent/configuration.py

"""Define the configurable parameters for the agent."""

from __future__ import annotations

from dataclasses import dataclass, fields
from typing import Optional

from langchain_core.runnables import RunnableConfig


@dataclass(kw_only=True)
class Configuration:


    model_name: str = "deepseek-chat"

    @classmethod
    def from_runnable_config(
        cls, config: Optional[RunnableConfig] = None
    ) -> Configuration:
        """Create a Configuration instance from a RunnableConfig object."""
        configurable = (config.get("configurable") or {}) if config else {}
        _fields = {f.name for f in fields(cls) if f.init}
        return cls(**{k: v for k, v in configurable.items() if k in _fields})

src/agent/graph.py



from typing import Any, Dict

from langchain_core.runnables import RunnableConfig
from langgraph.graph import StateGraph
from langchain_deepseek import ChatDeepSeek
from agent.configuration import Configuration
from agent.state import State

model = ChatDeepSeek(
    model="deepseek-chat",
    temperature=0,
    max_tokens=None,
    timeout=None,
    max_retries=2,
)

def call_model(state, config):
    response = model.invoke(state["messages"])
    # We return a list, because this will get added to the existing list
    return {"messages": [response]}


# Define a new graph
workflow = StateGraph(State, config_schema=Configuration)

# Add the node to the graph
workflow.add_node("call_model", call_model)

# Set the entrypoint as `call_model`
workflow.add_edge("__start__", "call_model")

# Compile the workflow into an executable graph
graph = workflow.compile()
graph.name = "New Graph"  # This defines the custom name in LangSmith

接下来就可以启动项目了：

# 安装依赖
pip install -qU langchain-deepseek -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install -e . -i https://pypi.tuna.tsinghua.edu.cn/simple

# 启动项目
langgraph dev

启动的运行日志如下图：
在这里插入图片描述
启动完毕后，会自动弹出浏览器打开langsmith跟踪页面，如下图：

API文档地址（http://127.0.0.1:2024/docs）访问如下图：
在这里插入图片描述

2 Agent Chat UI

前端项目基于NodeJS。

NodeJs版本 ≥ 18.12 我用的版本是 18.20.0

下载agent-chat-ui项目

git clone https://github.com/langchain-ai/agent-chat-ui.git

cd agent-chat-ui

# 先安装pnpm
npm install -g pnpm

# 安装项目依赖
pnpm install

# 启动项目
pnpm dev