34 lines
1.2 KiB
Python
34 lines
1.2 KiB
Python
"""
|
||
This code is supported by the website: https://www.guanjihuan.com
|
||
The newest version of this code is on the web page: https://www.guanjihuan.com/archives/47909
|
||
"""
|
||
|
||
from langchain_openai import ChatOpenAI
|
||
from langchain_core.prompts import ChatPromptTemplate
|
||
|
||
# 创建聊天模型 - 修改为调用 Ollama
|
||
llm = ChatOpenAI(
|
||
api_key="ollama", # 对于 Ollama,API key 可以设为任意值或 "ollama"
|
||
base_url="http://localhost:11434/v1", # Ollama 的本地 API 地址
|
||
model="qwen2.5:3b", # 替换为你本地安装的模型名称,如 qwen2.5 等
|
||
temperature=0.7, # 控制回复的随机性(0-1,越高越有创意)
|
||
streaming=True, # 启用流式模式
|
||
)
|
||
|
||
# 创建简单的提示词模板
|
||
prompt = ChatPromptTemplate.from_messages([
|
||
("system", "你是一个友好的聊天助手。"), # 系统角色设定
|
||
("human", "{question}") # 用户输入占位符
|
||
])
|
||
|
||
# 创建处理链
|
||
chain = prompt | llm # 使用管道操作符连接组件
|
||
|
||
# 使用 stream() 实现流式输出
|
||
for chunk in chain.stream({"question": "你好"}):
|
||
print(chunk.content, end="", flush=True)
|
||
print() # 换行
|
||
|
||
# # 非流式输出
|
||
# response = chain.invoke({"question": "你好"})
|
||
# print(response.content) |