> ## 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.

# Hugging Face 集成

> 使用 LangChain Python 与 Hugging Face 嵌入模型集成。

## 本地嵌入

你可以使用 `HuggingFaceEmbeddings` 类在本地生成嵌入。这利用了 `sentence_transformers` 库来下载模型权重，并直接在本地机器上运行。

让我们加载 Hugging Face 嵌入类。

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
pip install -qU  langchain langchain-huggingface sentence_transformers
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
from langchain_huggingface.embeddings import HuggingFaceEmbeddings
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-mpnet-base-v2")
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
text = "This is a test document."
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
query_result = embeddings.embed_query(text)
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
query_result[:3]
```

```text theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
[-0.04895168915390968, -0.03986193612217903, -0.021562768146395683]
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
doc_result = embeddings.embed_documents([text])
```

## Hugging Face 推理端点（无服务器 API）

如果你不希望本地下载模型，可以通过 [推理端点](https://huggingface.co/docs/inference-endpoints/index) 访问嵌入模型，这允许我们在 Hugging Face 的可扩展无服务器基础设施上使用开源模型。

确保已安装 huggingface\_hub，它通常随 langchain-huggingface 一起提供。

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
!pip install huggingface_hub
```

首先，我们需要从 [Hugging Face](https://huggingface.co/settings/tokens) 获取一个只读 API 密钥。

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
import os
from getpass import getpass

os.environ["HUGGINGFACEHUB_API_TOKEN"] = getpass()
```

现在我们可以使用 `HuggingFaceEndpointEmbeddings` 类通过 API 远程运行开源嵌入模型。

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
from langchain_huggingface.embeddings import HuggingFaceEndpointEmbeddings
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
embeddings = HuggingFaceEndpointEmbeddings(
    model="sentence-transformers/all-MiniLM-L6-v2"
)
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
text = "This is a test document."
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
query_result = embeddings.embed_query(text)
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
query_result[:3]
```

***

<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\embeddings\huggingfacehub.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>