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

# Cherry Studio integration

> Use Cherry Studio with Antigravity Manager for enhanced AI chat with multiple accounts

## Overview

Cherry Studio is a powerful AI chat client that supports multiple providers and advanced features. Antigravity Manager provides seamless integration with native support for web search citations and streaming responses.

<Info>
  Antigravity v4.1.21 introduced comprehensive Cherry Studio support, including SSE event completion and web search citation display.
</Info>

## Prerequisites

* Antigravity Manager installed and running
* Cherry Studio installed ([download here](https://github.com/kangfenmao/cherry-studio))
* At least one active account

## Configuration

<Steps>
  <Step title="Start Antigravity proxy">
    1. Open Antigravity Manager
    2. Navigate to **API Proxy** tab
    3. Enable the proxy server
    4. Verify port (default: 8045)
  </Step>

  <Step title="Add provider in Cherry Studio">
    1. Open Cherry Studio settings
    2. Go to **Providers** or **Models**
    3. Add new custom provider:
       * **Name**: Antigravity
       * **Base URL**: `http://127.0.0.1:8045/v1`
       * **API Key**: `sk-antigravity`
       * **Protocol**: OpenAI
  </Step>

  <Step title="Configure models">
    Add models you want to use:

    * `claude-sonnet-4-6`
    * `claude-sonnet-4-6-thinking`
    * `gemini-3-flash`
    * `gemini-3-pro-high`
    * `gemini-3-pro-image` (for image generation)
  </Step>
</Steps>

## Supported features

### Chat completions

Basic chat works out of the box:

```json theme={null}
{
  "model": "claude-sonnet-4-6",
  "messages": [
    {"role": "user", "content": "Hello!"}
  ]
}
```

Antigravity automatically:

* Routes to available accounts
* Handles streaming responses
* Manages quota protection

### Extended thinking

Use thinking models for complex reasoning:

<Tabs>
  <Tab title="Adaptive thinking">
    ```json theme={null}
    {
      "model": "claude-sonnet-4-6-thinking",
      "thinking": {
        "type": "adaptive"
      }
    }
    ```

    Antigravity maps this to 24576 token thinking budget.
  </Tab>

  <Tab title="Fixed budget">
    ```json theme={null}
    {
      "model": "claude-sonnet-4-6-thinking",
      "thinking": {
        "budget": 16384
      }
    }
    ```

    Explicitly set thinking token budget.
  </Tab>
</Tabs>

<Note>
  Antigravity v4.1.21+ automatically converts Cherry Studio's `thinking.type: "adaptive"` to a fixed budget compatible with Gemini's API.
</Note>

### Web search with citations

Cherry Studio's web search feature works seamlessly:

1. Enable web search in Cherry Studio chat
2. Antigravity automatically:
   * Injects Google Search tool
   * Executes search queries
   * Returns results with citations
   * Displays source links natively

**Example output:**

```markdown theme={null}
Based on my research:

**Source 1**: [Google Gemini API](https://ai.google.dev/api)
- Gemini 3 supports extended thinking mode

**Source 2**: [Anthropic Documentation](https://docs.anthropic.com)
- Claude models support adaptive reasoning
```

Cherry Studio will display these citations as clickable links.

### Image generation

Configure image generation models with custom parameters:

<Steps>
  <Step title="Add image model">
    In Cherry Studio, add model: `gemini-3-pro-image`
  </Step>

  <Step title="Configure image settings">
    Go to model settings in Cherry Studio:

    <ParamField path="size" type="string">
      Image dimensions, e.g., `1920x1080`, `1024x1024`, `16:9`

      **Default**: `1024x1024`
    </ParamField>

    <ParamField path="quality" type="string">
      Quality level: `standard`, `medium`, `hd`

      **Mapping**:

      * `standard` → 1K resolution
      * `medium` → 2K resolution
      * `hd` → 4K resolution
    </ParamField>

    <ParamField path="n" type="number">
      Number of images to generate (1-10)

      **Default**: 1
    </ParamField>
  </Step>

  <Step title="Generate images">
    Simply describe the image in chat:

    ```
    Generate: A futuristic cityscape at sunset
    ```

    Antigravity will:

    1. Route to image generation endpoint
    2. Apply configured size/quality
    3. Return image as Markdown
  </Step>
</Steps>

### Multi-modal input

Cherry Studio supports image input:

1. Click image attachment icon
2. Select image file
3. Add text prompt
4. Send message

Antigravity handles:

* Image encoding (Base64)
* Multi-modal request formatting
* Protocol conversion to Gemini

## Advanced configuration

### Model-specific settings

Configure per-model parameters in Cherry Studio:

<Tabs>
  <Tab title="Chat model">
    ```json theme={null}
    {
      "model": "claude-sonnet-4-6",
      "temperature": 0.7,
      "max_tokens": 4096,
      "top_p": 0.9
    }
    ```
  </Tab>

  <Tab title="Thinking model">
    ```json theme={null}
    {
      "model": "claude-sonnet-4-6-thinking",
      "temperature": 1.0,
      "thinking": {
        "type": "adaptive"
      },
      "max_tokens": 8192
    }
    ```
  </Tab>

  <Tab title="Image model">
    ```json theme={null}
    {
      "model": "gemini-3-pro-image",
      "size": "1920x1080",
      "quality": "hd",
      "n": 1
    }
    ```
  </Tab>
</Tabs>

### maxOutputTokens limit

Cherry Studio may send very large `maxOutputTokens` values (e.g., 128k) that exceed Gemini's limits:

<Warning>
  Antigravity automatically limits `maxOutputTokens` to **65536** for Claude protocol requests to prevent 400 errors.
</Warning>

This is handled automatically - no configuration needed.

## Troubleshooting

<AccordionGroup>
  <Accordion title="Empty responses with web search">
    **Problem**: Web search returns no content in Cherry Studio

    **Fixed in**: v4.1.21

    **Solution**:

    1. Upgrade Antigravity to v4.1.21 or later
    2. Restart both Antigravity and Cherry Studio
    3. Test web search again

    **Technical details**: Antigravity now sends complete SSE event lifecycle required by Cherry Studio.
  </Accordion>

  <Accordion title="Thinking mode 400 errors">
    **Problem**: `400 INVALID_ARGUMENT` with thinking models

    **Fixed in**: v4.1.21

    **Solution**:

    1. Ensure Antigravity v4.1.21+
    2. Use `claude-sonnet-4-6-thinking` model name
    3. Set thinking type to "adaptive" in Cherry Studio
    4. Antigravity automatically converts to compatible format
  </Accordion>

  <Accordion title="Connection errors">
    **Problem**: Cannot connect to Antigravity

    **Solutions**:

    1. Verify base URL: `http://127.0.0.1:8045/v1`
    2. Check Antigravity proxy is running
    3. Test with: `curl http://127.0.0.1:8045/health`
    4. Check firewall settings
  </Accordion>

  <Accordion title="Image generation errors">
    **Problem**: Image generation fails or times out

    **Solutions**:

    1. Check account has image generation quota
    2. Verify model name is `gemini-3-pro-image`
    3. Try smaller image sizes first
    4. Check Antigravity logs for detailed errors
    5. Ensure max body size is sufficient (100MB default)
  </Accordion>

  <Accordion title="Citations not appearing">
    **Problem**: Web search works but no citations shown

    **Solutions**:

    1. Verify Antigravity v4.1.21+
    2. Check Cherry Studio supports citation display
    3. Enable "Show sources" in Cherry Studio settings
    4. Review response format in debug logs
  </Accordion>
</AccordionGroup>

## Performance tips

<CardGroup cols={2}>
  <Card title="Model selection" icon="gauge-high">
    * **Fast responses**: Use `gemini-3-flash`
    * **Best quality**: Use `claude-sonnet-4-6`
    * **Complex reasoning**: Use thinking models
    * **Images**: Use `gemini-3-pro-image`
  </Card>

  <Card title="Quota management" icon="chart-line">
    * Monitor dashboard regularly
    * Enable quota protection
    * Use multiple accounts
    * Set appropriate model routing
  </Card>

  <Card title="Response speed" icon="bolt">
    * Keep Antigravity local (same machine)
    * Use streaming for long responses
    * Flash models for quick tasks
    * Reduce `max_tokens` when possible
  </Card>

  <Card title="Reliability" icon="shield">
    * Add multiple accounts for failover
    * Enable automatic retry
    * Configure rate limiting
    * Monitor error rates
  </Card>
</CardGroup>

## Best practices

1. **Use streaming**: Enable streaming in Cherry Studio for better UX
2. **Configure max tokens**: Set reasonable limits per model
3. **Enable web search selectively**: Only when needed to save quota
4. **Monitor quotas**: Check Antigravity dashboard regularly
5. **Test configurations**: Verify settings work before heavy use
6. **Keep updated**: Both Cherry Studio and Antigravity should be latest versions

## Example configurations

### Basic setup

```json theme={null}
{
  "provider": "Antigravity",
  "baseURL": "http://127.0.0.1:8045/v1",
  "apiKey": "sk-antigravity",
  "models": [
    {
      "name": "claude-sonnet-4-6",
      "maxTokens": 4096,
      "temperature": 0.7
    },
    {
      "name": "gemini-3-flash",
      "maxTokens": 8192,
      "temperature": 1.0
    }
  ]
}
```

### With image generation

```json theme={null}
{
  "models": [
    {
      "name": "gemini-3-pro-image",
      "type": "image",
      "size": "1920x1080",
      "quality": "hd"
    }
  ]
}
```

### With thinking models

```json theme={null}
{
  "models": [
    {
      "name": "claude-sonnet-4-6-thinking",
      "maxTokens": 8192,
      "thinking": {
        "type": "adaptive"
      }
    }
  ]
}
```

## Related documentation

* [OpenAI API format](/api/openai-format)
* [Image generation](/api/image-generation)
* [Model routing](/features/model-routing)
* [Quota monitoring](/features/quota-monitoring)
* [Streaming responses](/api/streaming)