> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/microsoft/graphrag/llms.txt
> Use this file to discover all available pages before exploring further.

# Multi-lingual support

> Use GraphRAG with documents in multiple languages

GraphRAG supports processing and querying documents in multiple languages by leveraging multilingual language models. This guide shows you how to configure GraphRAG for non-English content and mixed-language datasets.

## Language model selection

The key to multi-lingual support is choosing models that support your target languages:

<CardGroup cols={2}>
  <Card title="GPT-4 and GPT-4 Turbo" icon="comments">
    Excellent support for 50+ languages including European, Asian, and Middle Eastern languages
  </Card>

  <Card title="GPT-3.5 Turbo" icon="language">
    Good support for major languages, more cost-effective for large-scale processing
  </Card>

  <Card title="Multilingual embeddings" icon="chart-network">
    text-embedding-3-small and text-embedding-3-large support 100+ languages
  </Card>

  <Card title="Azure OpenAI" icon="cloud">
    Same language support with additional compliance and regional deployment options
  </Card>
</CardGroup>

## Basic configuration

No special configuration is needed for most languages. The default settings work well:

```yaml settings.yaml theme={null}
completion_models:
  default_completion_model:
    model_provider: openai
    model: gpt-4  # Supports multiple languages out of the box
    api_key: ${GRAPHRAG_API_KEY}

embedding_models:
  default_embedding_model:
    model_provider: openai
    model: text-embedding-3-small  # Multilingual embeddings
    api_key: ${GRAPHRAG_API_KEY}
```

## Language-specific prompt tuning

For better results, customize prompts in your target language:

<Tabs>
  <Tab title="Spanish">
    ```txt prompts/entity_extraction_es.txt theme={null}
    -Objetivo-
    Dado un documento de texto, identifique todas las entidades y sus relaciones.

    -Pasos-
    1. Identifique todas las entidades en el texto
    2. Para cada entidad, extraiga los atributos relevantes
    3. Identifique las relaciones entre entidades
    4. Formatee la salida según lo especificado

    -Tipos de Entidad-
    - PERSONA: Individuos humanos
    - ORGANIZACIÓN: Empresas, instituciones
    - UBICACIÓN: Lugares físicos o virtuales
    - EVENTO: Sucesos significativos
    ```
  </Tab>

  <Tab title="French">
    ```txt prompts/entity_extraction_fr.txt theme={null}
    -Objectif-
    Étant donné un document texte, identifiez toutes les entités et leurs relations.

    -Étapes-
    1. Identifiez toutes les entités dans le texte
    2. Pour chaque entité, extrayez les attributs pertinents
    3. Identifiez les relations entre les entités
    4. Formatez la sortie comme spécifié

    -Types d'Entité-
    - PERSONNE: Individus humains
    - ORGANISATION: Entreprises, institutions
    - LIEU: Endroits physiques ou virtuels
    - ÉVÉNEMENT: Événements importants
    ```
  </Tab>

  <Tab title="Japanese">
    ```txt prompts/entity_extraction_ja.txt theme={null}
    -目標-
    テキスト文書が与えられた場合、すべてのエンティティとその関係を識別します。

    -手順-
    1. テキスト内のすべてのエンティティを識別する
    2. 各エンティティについて、関連する属性を抽出する
    3. エンティティ間の関係を識別する
    4. 指定された形式で出力する

    -エンティティタイプ-
    - 人物: 人間個人
    - 組織: 企業、機関
    - 場所: 物理的または仮想的な場所
    - イベント: 重要な出来事
    ```
  </Tab>

  <Tab title="Chinese">
    ```txt prompts/entity_extraction_zh.txt theme={null}
    -目标-
    给定一个文本文档，识别所有实体及其关系。

    -步骤-
    1. 识别文本中的所有实体
    2. 对于每个实体，提取相关属性
    3. 识别实体之间的关系
    4. 按指定格式输出

    -实体类型-
    - 人物: 人类个体
    - 组织: 公司、机构
    - 地点: 物理或虚拟位置
    - 事件: 重要事件
    ```
  </Tab>
</Tabs>

## Processing mixed-language documents

When working with documents in multiple languages:

<Steps>
  <Step title="Organize by language">
    Optionally separate documents by language for better tracking:

    ```bash theme={null}
    input/
    ├── en/
    │   ├── document1.txt
    │   └── document2.txt
    ├── es/
    │   ├── documento1.txt
    │   └── documento2.txt
    └── fr/
        ├── document1.txt
        └── document2.txt
    ```
  </Step>

  <Step title="Use language-agnostic prompts">
    For truly mixed content, use English prompts with instructions to handle multiple languages:

    ```txt theme={null}
    -Goal-
    Given a text document that may contain content in multiple languages,
    identify all entities and relationships. Preserve entity names in their
    original language and provide descriptions in the source language.

    -Instructions-
    - Maintain entity names in original language
    - Generate descriptions in the same language as the source text
    - Identify language of each extracted entity
    ```
  </Step>

  <Step title="Configure language metadata">
    Track language information in your chunking configuration:

    ```yaml theme={null}
    chunking:
      prepend_metadata: ["language", "source_file"]
    ```
  </Step>
</Steps>

## Querying in multiple languages

GraphRAG supports queries in different languages:

<CodeGroup>
  ```bash English theme={null}
  graphrag query "What are the main themes in this dataset?"
  ```

  ```bash Spanish theme={null}
  graphrag query "¿Cuáles son los temas principales en este conjunto de datos?"
  ```

  ```bash French theme={null}
  graphrag query "Quels sont les thèmes principaux de cet ensemble de données?"
  ```

  ```bash Japanese theme={null}
  graphrag query "このデータセットの主なテーマは何ですか？"
  ```

  ```bash Chinese theme={null}
  graphrag query "这个数据集的主要主题是什么？"
  ```
</CodeGroup>

<Note>
  The LLM will attempt to respond in the same language as your query. For mixed-language datasets, you can specify the desired response language in your query.
</Note>

## Best practices for specific languages

<AccordionGroup>
  <Accordion title="Chinese (Simplified/Traditional)">
    * Use larger chunk sizes due to character density
    * Consider using `gpt-4` for better understanding of classical vs. modern Chinese
    * Test entity extraction with both simplified and traditional characters

    ```yaml theme={null}
    chunking:
      size: 800  # Increase from default 400
      overlap: 200  # Increase overlap proportionally
    ```
  </Accordion>

  <Accordion title="Japanese">
    * Account for mixed scripts (Hiragana, Katakana, Kanji)
    * Use character-based rather than token-based chunking
    * Test entity extraction with company names (often use Kanji)

    ```yaml theme={null}
    chunking:
      type: sentence  # Better for Japanese text
      size: 600
    ```
  </Accordion>

  <Accordion title="Arabic/Hebrew (RTL languages)">
    * Ensure text encoding is UTF-8
    * Be aware that some entity names may be transliterated
    * Test with mixed RTL/LTR content (common in technical documents)

    ```yaml theme={null}
    input:
      encoding: utf-8
    ```
  </Accordion>

  <Accordion title="German">
    * Long compound words may need special handling
    * Consider noun capitalization in entity extraction
    * Adjust chunk sizes for longer words

    ```yaml theme={null}
    chunking:
      size: 500  # Slightly larger for compound words
    ```
  </Accordion>
</AccordionGroup>

## Example: Multi-lingual research corpus

Here's a complete example for processing academic papers in multiple languages:

<Steps>
  <Step title="Prepare data">
    Organize papers with language metadata:

    ```csv input/papers.csv theme={null}
    id,title,text,language,author,year
    1,"Machine Learning Basics","Full text...",en,"Smith",2023
    2,"Apprentissage Automatique","Texte complet...",fr,"Dubois",2023
    3,"機械学習の基礎","全文...",ja,"田中",2023
    ```
  </Step>

  <Step title="Configure for CSV input">
    ```yaml settings.yaml theme={null}
    input:
      type: csv
      file_pattern: .*\.csv$
      id_column: id
      title_column: title
      text_column: text

    chunking:
      prepend_metadata: ["language", "author", "year"]
      size: 600
      overlap: 100
    ```
  </Step>

  <Step title="Create multilingual prompts">
    ```txt prompts/entity_extraction_multilingual.txt theme={null}
    -Goal-
    Extract academic entities from research papers in any language.
    Preserve technical terms and names in their original language.

    -Entity Types-
    - RESEARCHER: Authors and cited researchers
    - CONCEPT: Scientific concepts and methods
    - INSTITUTION: Universities and research organizations
    - PUBLICATION: Papers, journals, conferences

    -Instructions-
    - Keep researcher names in original form
    - Preserve technical terminology in source language
    - Translate descriptions to match source document language
    ```
  </Step>

  <Step title="Query across languages">
    ```bash theme={null}
    # Query in English
    graphrag query "What machine learning concepts are discussed?" --method global

    # Query in French
    graphrag query "Quels chercheurs sont mentionnés?" --method local

    # Query in Japanese  
    graphrag query "どの研究機関が関与していますか？" --method local
    ```
  </Step>
</Steps>

## Language detection and routing

For advanced use cases, implement language detection:

```python theme={null}
from langdetect import detect
import pandas as pd

# Detect language in documents
df = pd.read_csv('input/documents.csv')
df['detected_language'] = df['text'].apply(detect)

# Route to language-specific prompts
language_prompts = {
    'en': 'prompts/entity_extraction_en.txt',
    'es': 'prompts/entity_extraction_es.txt',
    'fr': 'prompts/entity_extraction_fr.txt',
    'ja': 'prompts/entity_extraction_ja.txt',
}
```

## Troubleshooting

<AccordionGroup>
  <Accordion title="Poor entity extraction in target language">
    Solutions:

    * Use language-specific prompts
    * Run auto prompt tuning with documents in target language
    * Increase chunk size for languages with longer words/characters
    * Verify model supports your target language well
  </Accordion>

  <Accordion title="Mixed language responses">
    Solutions:

    * Explicitly specify response language in your query
    * Use language-specific system prompts
    * Separate documents by language during indexing
  </Accordion>

  <Accordion title="Encoding issues">
    Solutions:

    * Ensure all files are UTF-8 encoded
    * Verify `.env` file doesn't have encoding issues
    * Check that storage systems support Unicode
  </Accordion>
</AccordionGroup>

## Supported languages

OpenAI models (GPT-4, GPT-3.5-turbo, embeddings) have strong support for:

<Tabs>
  <Tab title="European">
    * English, Spanish, French, German, Italian
    * Portuguese, Dutch, Polish, Russian
    * Swedish, Norwegian, Danish, Finnish
    * Greek, Turkish, Czech, Romanian
  </Tab>

  <Tab title="Asian">
    * Chinese (Simplified & Traditional)
    * Japanese, Korean
    * Vietnamese, Thai, Indonesian
    * Hindi, Bengali, Tamil
  </Tab>

  <Tab title="Middle Eastern">
    * Arabic, Hebrew, Persian (Farsi)
    * Turkish, Urdu
  </Tab>

  <Tab title="Others">
    * Swahili, Zulu
    * Tagalog, Malay
    * Over 100 total languages supported
  </Tab>
</Tabs>

## Next steps

<CardGroup cols={2}>
  <Card title="Custom prompts" icon="pen" href="/examples/custom-prompts">
    Create language-specific prompts
  </Card>

  <Card title="Document Q&A" icon="comments" href="/examples/use-cases/document-qa">
    Build multilingual Q\&A systems
  </Card>

  <Card title="Azure deployment" icon="cloud" href="/examples/azure-deployment">
    Deploy with regional Azure endpoints
  </Card>

  <Card title="Prompt tuning" icon="wand-magic-sparkles" href="/prompt-tuning/auto-tuning">
    Auto-tune prompts for your language
  </Card>
</CardGroup>