> ## Documentation Index
> Fetch the complete documentation index at: https://langchain-zh.cn/llms.txt
> Use this file to discover all available pages before exploring further.
# Cohere 集成
> 使用 LangChain Python 与 Cohere 集成。
> [Cohere](https://cohere.ai/about) 是一家加拿大初创公司，提供自然语言处理模型，帮助企业改善人机交互。
## 安装和设置
* 安装 Python SDK：
```bash pip theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
pip install langchain-cohere
```
```bash uv theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
uv add langchain-cohere
```
获取 [Cohere API 密钥](https://dashboard.cohere.ai/) 并将其设置为环境变量 (`COHERE_API_KEY`)
## Cohere LangChain 集成
| API | 描述 | 端点文档 | 导入 | 示例用法 |
| ------- | ------------- | -------------------------------------------------- | ---------------------------------------------------------------------------- | ------------------------------------------------------------------- |
| 聊天 | 构建聊天机器人 | [聊天](https://docs.cohere.com/reference/chat) | `from langchain_cohere import ChatCohere` | [cohere.ipynb](/oss/python/integrations/chat/cohere) |
| LLM | 生成文本 | [生成](https://docs.cohere.com/reference/generate) | `from langchain_cohere.llms import Cohere` | [cohere.ipynb](/oss/python/integrations/llms/cohere) |
| RAG 检索器 | 连接到外部数据源 | [聊天 + RAG](https://docs.cohere.com/reference/chat) | `from langchain_classic.retrievers import CohereRagRetriever` | [cohere.ipynb](/oss/python/integrations/retrievers/cohere) |
| 文本嵌入 | 将字符串嵌入为向量 | [嵌入](https://docs.cohere.com/reference/embed) | `from langchain_cohere import CohereEmbeddings` | [cohere.ipynb](/oss/python/integrations/embeddings/cohere) |
| 重排序检索器 | 根据相关性对字符串进行排名 | [重排序](https://docs.cohere.com/reference/rerank) | `from langchain_classic.retrievers.document_compressors import CohereRerank` | [cohere.ipynb](/oss/python/integrations/retrievers/cohere-reranker) |
## 快速复制示例
### 聊天
```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
from langchain_cohere import ChatCohere
from langchain.messages import HumanMessage
chat = ChatCohere()
messages = [HumanMessage(content="knock knock")]
print(chat.invoke(messages))
```
Cohere [聊天模型](/oss/python/integrations/chat/cohere) 的用法
### LLM
```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
from langchain_cohere.llms import Cohere
llm = Cohere()
print(llm.invoke("Come up with a pet name"))
```
Cohere (遗留) [LLM 模型](/oss/python/integrations/llms/cohere) 的用法
### 工具调用
```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
from langchain_cohere import ChatCohere
from langchain.messages import (
HumanMessage,
ToolMessage,
)
from langchain.tools import tool
@tool
def magic_function(number: int) -> int:
"""Applies a magic operation to an integer
Args:
number: Number to have magic operation performed on
"""
return number + 10
def invoke_tools(tool_calls, messages):
for tool_call in tool_calls:
selected_tool = {"magic_function":magic_function}[
tool_call["name"].lower()
]
tool_output = selected_tool.invoke(tool_call["args"])
messages.append(ToolMessage(tool_output, tool_call_id=tool_call["id"]))
return messages
tools = [magic_function]
llm = ChatCohere()
llm_with_tools = llm.bind_tools(tools=tools)
messages = [
HumanMessage(
content="What is the value of magic_function(2)?"
)
]
res = llm_with_tools.invoke(messages)
while res.tool_calls:
messages.append(res)
messages = invoke_tools(res.tool_calls, messages)
res = llm_with_tools.invoke(messages)
print(res.content)
```
Cohere LLM 的工具调用可以通过如上所示将必要的工具绑定到 llm 来完成。
另一种选择是使用 ReAct 代理支持多跳工具调用，如下所示。
### ReAct 代理
该代理基于论文
[ReAct: Synergizing Reasoning and Acting in Language Models](https://arxiv.org/abs/2210.03629)。
```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
from langchain_community.tools.tavily_search import TavilySearchResults
from langchain_cohere import ChatCohere, create_cohere_react_agent
from langchain_core.prompts import ChatPromptTemplate
from langchain.agents import AgentExecutor
llm = ChatCohere()
internet_search = TavilySearchResults(max_results=4)
internet_search.name = "internet_search"
internet_search.description = "Route a user query to the internet"
prompt = ChatPromptTemplate.from_template("{input}")
agent = create_cohere_react_agent(
llm,
[internet_search],
prompt
)
agent_executor = AgentExecutor(agent=agent, tools=[internet_search], verbose=True)
agent_executor.invoke({
"input": "In what year was the company that was founded as Sound of Music added to the S&P 500?",
})
```
ReAct 代理可用于按顺序调用多个工具。
### RAG 检索器
```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
from langchain_cohere import ChatCohere
from langchain_classic.retrievers import CohereRagRetriever
from langchain_core.documents import Document
rag = CohereRagRetriever(llm=ChatCohere())
print(rag.invoke("What is cohere ai?"))
```
Cohere [RAG 检索器](/oss/python/integrations/retrievers/cohere) 的用法
### 文本嵌入
```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
from langchain_cohere import CohereEmbeddings
embeddings = CohereEmbeddings(model="embed-english-light-v3.0")
print(embeddings.embed_documents(["This is a test document."]))
```
Cohere [文本嵌入模型](/oss/python/integrations/embeddings/cohere) 的用法
### 重排序器
Cohere [重排序器](/oss/python/integrations/retrievers/cohere-reranker) 的用法
***
[Edit this page on GitHub](https://github.com/langchain-ai/docs/edit/main/src/i18n\zh-CN\oss\python\integrations\providers\cohere.mdx) or [file an issue](https://github.com/langchain-ai/docs/issues/new/choose).
[Connect these docs](/use-these-docs) to Claude, VSCode, and more via MCP for real-time answers.