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

# Cherry Studio

> Cherry Studio Client Integration Guide

## Introduction

Cherry Studio is a powerful AI chat client that supports multiple large language models. With **GravitexAI**, you can easily use various mainstream AI models in Cherry Studio.

## Quick Integration

### 1. Get API Key

Visit [GravitexAI](https://maas.gravitex.ai) to get your API key.

### 2. Configuration

1. Open Cherry Studio
2. Click the settings icon on the left
3. Select "Model Service"
4. Fill in:
   * API URL: `https://api.gravitex.ai`
   * API Key: Your key from GravitexAI console
5. Click "➕ Add" to save

<Warning>
  API URL must be `https://api.gravitex.ai`
</Warning>

### 3. Add Models

1. Search for target model (e.g. `gemini-3.1-pro-preview`)
2. Click the icon next to model name to configure
3. Enable/disable model variants as needed
4. Use right-side buttons for detailed configuration

## Supported Models

### Latest Recommended

| Series            | Model ID                 | Features                                                               |
| ----------------- | ------------------------ | ---------------------------------------------------------------------- |
| Gemini 3.1 Pro    | `gemini-3.1-pro-preview` | Google's upgraded core intelligence model for complex multimodal tasks |
| GPT-5.4           | `gpt-5.4`                | OpenAI's latest mainline model with strong overall performance         |
| Claude Sonnet 4.6 | `claude-sonnet-4-6`      | Anthropic's latest Sonnet model for long-text and coding tasks         |
| DeepSeek V3.2     | `deepseek-v3-2-251201`   | High-value hybrid-reasoning model for general workloads                |

### Classic Stable

| Series                | Model ID                        | Use Case                                              |
| --------------------- | ------------------------------- | ----------------------------------------------------- |
| Claude Opus 4.6       | `claude-opus-4-6`               | Complex reasoning, code review, high-difficulty tasks |
| Gemini 3.1 Flash Lite | `gemini-3.1-flash-lite-preview` | Fast response, low-cost high-concurrency chat         |

### Selection Guide

* **Multimodal**: `gemini-3.1-pro-preview`
* **Daily Chat**: `gpt-5.4` or `gemini-3.1-flash-lite-preview`
* **Long Text**: `claude-sonnet-4-6` or `claude-opus-4-6`
* **Complex Reasoning**: `gpt-5.4-pro`, `claude-opus-4-6`, `deepseek-r1-250528`

## Advanced Features

### Image Support

For vision-enabled models (e.g. `GPT-4V`):

1. Enable "Images" in settings
2. Select a vision-capable model
3. Upload images directly in chat

### Streaming Output

Cherry Studio supports streaming output by default for smoother conversations.

## Troubleshooting

<AccordionGroup>
  <Accordion title="Connection Failed">
    * Verify API key is correct
    * Confirm API URL is `https://api.gravitex.ai`
    * Check network connection
  </Accordion>

  <Accordion title="Model Unavailable">
    * Confirm sufficient account balance
    * Check if model is in subscription plan
    * Try switching to another model
  </Accordion>
</AccordionGroup>

## Tips

1. **Choose wisely**: Match models to task requirements
2. **Stay updated**: Follow new model releases
3. **Monitor usage**: Check usage via GravitexAI console