> ## Documentation Index
> Fetch the complete documentation index at: https://docs.laozhang.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Immersive Translate Extension

> Immersive Translate bilingual web page translation extension AI model integration guide

Immersive Translate is a powerful bilingual web page translation browser extension. Through Laozhang API, you can integrate various advanced AI models into Immersive Translate for a more natural and accurate translation experience.

## Quick Integration

### Step 1: Install Extension

Install Immersive Translate from the following browsers:

* **Chrome/Edge**: [Chrome Web Store](https://chrome.google.com/webstore/detail/immersive-translate/bpoadfkcbjbfhfodiogcnhhhpibjhbnh)
* **Firefox**: [Firefox Add-ons](https://addons.mozilla.org/firefox/addon/immersive-translate/)
* **Safari**: [Safari Extensions](https://apps.apple.com/app/immersive-translate/id6447957425)

### Step 2: Configure Laozhang API

<Steps>
  <Step title="Open Settings">
    Click the Immersive Translate extension icon, select "Settings"
  </Step>

  <Step title="Choose Translation Service">
    In "Translation Services", select "OpenAI"
  </Step>

  <Step title="Fill In API Configuration">
    Configure the following information:

    * **API Key**: Enter your Laozhang API key
    * **Custom API URL**: `https://api2.laozhang.ai/v1/chat/completions`
    * **Model**: Select the model you need (e.g., `gpt-4-turbo`)
  </Step>

  <Step title="Test Configuration">
    Click "Test" button to ensure connection is successful
  </Step>

  <Step title="Save Settings">
    Click "Save" button to complete configuration
  </Step>
</Steps>

<Info>
  **Obtaining API Key**

  Visit [Laozhang API Console](https://api2.laozhang.ai/token) to create and obtain your API key.
</Info>

## Recommended Model Selection

Choose appropriate models based on different translation scenarios:

| Translation Scenario        | Recommended Model | Speed  | Quality   | Cost   |
| --------------------------- | ----------------- | ------ | --------- | ------ |
| **General Web Translation** | `gpt-3.5-turbo`   | Fast   | Good      | Low    |
| **Professional Articles**   | `gpt-4-turbo`     | Medium | Excellent | Medium |
| **Technical Documentation** | `gpt-4-turbo`     | Medium | Excellent | Medium |
| **News Reading**            | `gpt-3.5-turbo`   | Fast   | Good      | Low    |
| **Academic Papers**         | `claude-sonnet-4` | Medium | Excellent | Medium |
| **Literary Works**          | `claude-sonnet-4` | Medium | Excellent | Medium |

<Tip>
  **Model Selection Recommendations**

  * **Daily browsing**: Use `gpt-3.5-turbo`, fast and economical
  * **Professional content**: Use `gpt-4-turbo`, high accuracy
  * **Long articles**: Use `claude-sonnet-4`, supports long context
  * **Technical documents**: Use `gpt-4-turbo`, precise terminology
</Tip>

## Core Features

### 1. Bilingual Comparison Translation

Display original and translation side-by-side:

* **Paragraph-by-paragraph correspondence**: Clear and intuitive
* **Hover highlighting**: Highlight corresponding paragraphs
* **Quick positioning**: Quickly find target content

### 2. Hover Translation

Hover over words for quick translation:

* No need to copy and paste
* Instant results
* Supports multiple languages

### 3. Input Box Translation

Translate while typing:

* Real-time translation feedback
* Supports multiple platforms
* Quick bilingual switching

### 4. PDF Translation

Translate PDF documents:

* Maintain original layout
* Support bilingual comparison
* Export translated files

### 5. Subtitle Translation

Translate video subtitles:

* Support YouTube, Netflix, etc.
* Real-time subtitle translation
* Bilingual subtitle display

## Usage Scenarios

### Scenario 1: Reading Technical Documentation

Reading technical articles/documentation on GitHub, Medium, etc.:

1. Open the page you need to read
2. Click the Immersive Translate icon
3. Automatically translate to bilingual display
4. Hover over important paragraphs for further reading

### Scenario 2: Academic Research

Reading English academic papers:

1. Open the paper page
2. Enable translation feature
3. Bilingual comparison reading
4. Accurately understand terminology
5. Take notes and cite

### Scenario 3: News Browsing

Reading international news:

1. Visit foreign news websites
2. Automatically trigger translation
3. Quickly browse headlines
4. In-depth reading of interesting content

### Scenario 4: Video Learning

Watching English tutorial videos:

1. Open YouTube/Coursera videos
2. Enable subtitle translation
3. Real-time bilingual subtitle display
4. Better understand content

## Advanced Configuration

### Custom Prompts

Customize prompts for translation:

```
You are a professional translator proficient in English and Chinese.

Task: Translate the following text to Chinese
Requirements:
1. Accurate and fluent translation
2. Maintain professional terminology accuracy
3. Keep paragraph structure unchanged
4. Ensure text is natural and fluent

Text to translate:
{text}

Please provide only the translation, without explanations.
```

### Translation Style Settings

Set different translation styles:

| Style        | Temperature | Features             | Use Cases                        |
| ------------ | ----------- | -------------------- | -------------------------------- |
| **Literal**  | 0.1         | Precise and accurate | Technical documents, contracts   |
| **Balanced** | 0.3         | Accurate and fluent  | News, general articles           |
| **Free**     | 0.5         | Natural and fluent   | Literary works, blogs            |
| **Creative** | 0.7         | Creative and diverse | Marketing copy, creative content |

### Domain-specific Configuration

Optimize translation for specific domains:

```json theme={null}
{
  "tech": {
    "model": "gpt-4-turbo",
    "temperature": 0.2,
    "systemPrompt": "You are a technical document translator, proficient in various programming languages and technical terms."
  },
  "medical": {
    "model": "gpt-4-turbo",
    "temperature": 0.1,
    "systemPrompt": "You are a medical document translator, proficient in medical terminology and professional expressions."
  },
  "legal": {
    "model": "claude-sonnet-4",
    "temperature": 0.1,
    "systemPrompt": "You are a legal document translator, proficient in legal terminology and formal expressions."
  }
}
```

## Feature Tips

### Shortcut Keys

| Shortcut  | Function                 |
| --------- | ------------------------ |
| `Alt + A` | Toggle translation       |
| `Alt + W` | Translate entire page    |
| `Alt + S` | Show/hide original text  |
| `Alt + D` | Toggle translation style |

### Smart Recognition

Auto-recognize and adapt:

* **Page Language**: Auto-detect source language
* **Content Type**: Auto-adjust translation strategy
* **User Preferences**: Learn your habits

### Cache Mechanism

Improve efficiency:

* **Translation Cache**: Repeat content no re-translation
* **Terminology Cache**: Consistent professional terms
* **Speed Optimization**: Faster page loading

## Troubleshooting

### Connection Issues

**Problem**: Unable to connect to API

**Solutions**:

1. Check if custom API URL is correct: `https://api2.laozhang.ai/v1/chat/completions`
2. Verify API Key validity
3. Check network connection
4. Confirm firewall/proxy settings

### Translation Quality Poor

**Problem**: Translation results don't meet expectations

**Solutions**:

1. Switch to a more advanced model (e.g., from GPT-3.5 to GPT-4)
2. Adjust translation style (decrease temperature)
3. Optimize prompt template
4. Check if source language is correctly identified

### Slow Translation Speed

**Problem**: Translation response is slow

**Solutions**:

1. Switch to a faster model (e.g., `gpt-3.5-turbo`)
2. Enable cache feature
3. Check network speed
4. Reduce max\_tokens setting

### Some Content Not Translated

**Problem**: Some content on page is not translated

**Solutions**:

1. Check if excluded translation selectors are set
2. Refresh page and retry
3. Manually trigger translation
4. Check page structure if special

## Best Practices

### 1. Choose Models Wisely

Choose appropriate models based on scenarios:

```
Daily browsing → gpt-3.5-turbo
    ↓
Important articles → gpt-4-turbo
    ↓
Professional papers → claude-sonnet-4
```

### 2. Manage Translation Rules

Set translation rules:

* **Never translate**: Set domain names or page selectors not to translate
* **Always translate**: Set auto-translation domain names
* **Translation style**: Set default style for different websites

### 3. Optimize Reading Experience

```
1. Enable bilingual comparison → Easy to verify accuracy
2. Use hover translation → Quick understanding of terms
3. Enable cache → Improve speed
4. Customize shortcuts → More efficient
```

### 4. Cost Control

Control usage costs:

* Use `gpt-3.5-turbo` for daily browsing
* Use advanced models only for important content
* Enable cache to reduce repeat translations
* Set daily usage limit

## Plugin Configuration Examples

### Daily Browsing Configuration

```json theme={null}
{
  "model": "gpt-3.5-turbo",
  "temperature": 0.3,
  "maxTokens": 1000,
  "cacheEnabled": true,
  "autoTranslate": false
}
```

### Professional Reading Configuration

```json theme={null}
{
  "model": "gpt-4-turbo",
  "temperature": 0.2,
  "maxTokens": 2000,
  "cacheEnabled": true,
  "autoTranslate": true,
  "bilingualMode": true
}
```

### Cost-optimized Configuration

```json theme={null}
{
  "model": "gpt-3.5-turbo",
  "temperature": 0.3,
  "maxTokens": 800,
  "cacheEnabled": true,
  "autoTranslate": false,
  "onlyTranslateSelection": true
}
```

## Cost Optimization Strategy

### Usage Strategy

| Content Type              | Model   | Max Tokens | Expected Cost |
| ------------------------- | ------- | ---------- | ------------- |
| **News Headlines**        | GPT-3.5 | 500        | Very Low      |
| **General Articles**      | GPT-3.5 | 1000       | Low           |
| **Professional Articles** | GPT-4   | 1500       | Medium        |
| **Academic Papers**       | Claude  | 2000       | High          |

### Saving Tips

1. **Enable Cache**: Avoid repeat translations
2. **Selective Translation**: Translate only what you need
3. **Batch Processing**: Translate multiple paragraphs at once
4. **Model Switching**: Use appropriate models for different content
5. **Limit Token Count**: Set reasonable max\_tokens

## Further Learning

### Official Resources

* [Immersive Translate Official Documentation](https://immersivetranslate.com/docs/)
* [Immersive Translate GitHub](https://github.com/immersive-translate/immersive-translate)
* [Immersive Translate Community](https://github.com/immersive-translate/immersive-translate/discussions)

### Learning Resources

* [Quick Start Tutorial](https://immersivetranslate.com/docs/quickstart/)
* [Advanced Configuration Guide](https://immersivetranslate.com/docs/advanced/)
* [FAQ](https://immersivetranslate.com/docs/faq/)

### Community Support

* [GitHub Issues](https://github.com/immersive-translate/immersive-translate/issues)
* [User Forum](https://github.com/immersive-translate/immersive-translate/discussions)
* [Twitter](https://twitter.com/immersivetran)

Need more help? Please visit [Laozhang API Official Website](https://api2.laozhang.ai) or contact our support team.