> ## 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.

# Together AI 集成

> 使用 LangChain Python 与 Together AI 集成。

[Together AI](https://www.together.ai/) 提供了一个 API，只需几行代码即可查询 [50+ leading open-source models](https://docs.together.ai/docs/inference-models)。

本示例介绍如何使用 LangChain 与 Together AI 模型进行交互。

## 安装

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
pip install -U langchain-together
```

## 环境

要使用 Together AI，您需要一个 API 密钥，可以在这里找到：
[api.together.ai/settings/api-keys](https://api.together.ai/settings/api-keys)。这可以作为初始化参数
`together_api_key` 或设置为环境变量 `TOGETHER_API_KEY`。

## 示例

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
# Querying chat models with Together AI

from langchain_together import ChatTogether

# choose from our 50+ models here: https://docs.together.ai/docs/inference-models
chat = ChatTogether(
    # together_api_key="YOUR_API_KEY",
    model="meta-llama/Llama-3-70b-chat-hf",
)

# stream the response back from the model
for m in chat.stream("Tell me fun things to do in NYC"):
    print(m.content, end="", flush=True)

# if you don't want to do streaming, you can use the invoke method
# chat.invoke("Tell me fun things to do in NYC")
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
# Querying code and language models with Together AI

from langchain_together import Together

llm = Together(
    model="codellama/CodeLlama-70b-Python-hf",
    # together_api_key="..."
)

print(llm.invoke("def bubble_sort(): "))
```

***

<div className="source-links">
  <Callout icon="edit">
    [Edit this page on GitHub](https://github.com/langchain-ai/docs/edit/main/src/i18n\zh-CN\oss\python\integrations\providers\together.mdx) or [file an issue](https://github.com/langchain-ai/docs/issues/new/choose).
  </Callout>

  <Callout icon="terminal-2">
    [Connect these docs](/use-these-docs) to Claude, VSCode, and more via MCP for real-time answers.
  </Callout>
</div>