> ## Documentation Index
> Fetch the complete documentation index at: https://docs.comput3.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# VSCode Setup

> Configure Visual Studio Code to use Comput3 Network for AI assistance

Configure Visual Studio Code with AI extensions to use the Comput3 Network API for enhanced coding assistance.

## Prerequisites

* Visual Studio Code installed ([Download here](https://code.visualstudio.com))
* Comput3 API key ([Get yours here](https://app.comput3.ai))

## Recommended Extensions

<CardGroup cols={2}>
  <Card title="GitHub Copilot" icon="github">
    **Alternative**: Use Comput3 as OpenAI-compatible backend
  </Card>

  <Card title="CodeGPT" icon="robot">
    **Recommended**: Native support for custom API endpoints
  </Card>

  <Card title="Continue" icon="arrow-right">
    **Best Choice**: Open-source with full Comput3 support
  </Card>

  <Card title="Tabnine" icon="tab">
    **Enterprise**: Custom model deployment options
  </Card>
</CardGroup>

## Setup with Continue Extension (Recommended)

<Steps>
  <Step title="Install Continue Extension">
    1. Open VSCode Extensions panel (`Cmd+Shift+X` or `Ctrl+Shift+X`)
    2. Search for "Continue"
    3. Install the Continue extension by Continue Dev

    <Frame>
      <img src="https://mintlify.s3.us-west-1.amazonaws.com/comput3-e7edbb9d/images/vscode-continue.png" alt="Continue extension in VSCode marketplace" />
    </Frame>
  </Step>

  <Step title="Configure Comput3 API">
    1. Open Continue configuration (`Cmd+Shift+P` → "Continue: Open config.json")
    2. Replace the configuration with:

    ```json config.json theme={null}
    {
      "models": [
        {
          "title": "Comput3 DeepSeek",
          "provider": "openai",
          "model": "hermes4:70b",
          "apiKey": "YOUR_COMPUT3_API_KEY",
          "apiBase": "https://api.comput3.ai/v1"
        },
        {
          "title": "Comput3 Hermes",
          "provider": "openai",
          "model": "hermes4:70b",
          "apiKey": "YOUR_COMPUT3_API_KEY", 
          "apiBase": "https://api.comput3.ai/v1"
        }
      ],
      "tabAutocompleteModel": {
        "title": "Comput3 Autocomplete",
        "provider": "openai",
        "model": "kimi-k2",
        "apiKey": "YOUR_COMPUT3_API_KEY",
        "apiBase": "https://api.comput3.ai/v1"
      },
      "allowAnonymousTelemetry": false
    }
    ```

    <Warning>
      Replace `YOUR_COMPUT3_API_KEY` with your actual API key from the Comput3 dashboard.
    </Warning>
  </Step>

  <Step title="Test the Setup">
    1. Open any code file in VSCode
    2. Highlight some code and press `Cmd+Shift+M` (or `Ctrl+Shift+M`)
    3. Ask "Explain this code" or request a code review

    <Check>
      You should see responses powered by Comput3's GPU infrastructure.
    </Check>
  </Step>
</Steps>

## Setup with CodeGPT Extension

<Steps>
  <Step title="Install CodeGPT">
    1. Search for "CodeGPT" in VSCode Extensions
    2. Install the CodeGPT extension
    3. Reload VSCode
  </Step>

  <Step title="Configure API Settings">
    1. Open Settings (`Cmd+,` or `Ctrl+,`)
    2. Search for "CodeGPT"
    3. Set the following configurations:

    <ParamField path="API Provider" type="string" required>
      Select "OpenAI Compatible"
    </ParamField>

    <ParamField path="API Base URL" type="string" required>
      ```
      https://api.comput3.ai/v1
      ```
    </ParamField>

    <ParamField path="API Key" type="string" required>
      Your Comput3 API key
    </ParamField>

    <ParamField path="Model" type="string" default="hermes4:70b">
      ```
      hermes4:70b
      hermes4:405b
      deepseek-v3.1
      kimi-k2
      qwen3-coder:480b
      qwen3-max
      grok-code-fast-1
      claude-sonnet-4
      ```
    </ParamField>
  </Step>
</Steps>

## Environment Variables Setup

For team projects, use environment variables:

<Tabs>
  <Tab title="macOS/Linux">
    Add to your shell profile (`~/.bashrc`, `~/.zshrc`):

    ```bash theme={null}
    export COMPUT3_API_KEY="your_api_key_here"
    export OPENAI_API_BASE="https://api.comput3.ai/v1"
    ```
  </Tab>

  <Tab title="Windows">
    Set in PowerShell or System Environment Variables:

    ```powershell theme={null}
    $env:COMPUT3_API_KEY="your_api_key_here"
    $env:OPENAI_API_BASE="https://api.comput3.ai/v1"
    ```
  </Tab>
</Tabs>

## Features and Usage

### Code Completion

* **Trigger**: Type code and pause for auto-suggestions
* **Accept**: Press `Tab` to accept suggestions
* **Reject**: Press `Esc` or continue typing

### Code Chat

* **Trigger**: `Cmd+Shift+P` → "CodeGPT: Ask"
* **Usage**: Ask questions about your code, request explanations, or get help

### Code Generation

* **Trigger**: Write a comment describing what you want
* **Example**: `// Create a function to validate email addresses`
* **Generate**: Press `Cmd+Enter` or use the command palette

### Code Review

* **Select**: Highlight code you want reviewed
* **Review**: Right-click → "CodeGPT: Review Code"
* **Result**: Get suggestions for improvements, bugs, and optimizations

## Troubleshooting

<AccordionGroup>
  <Accordion title="Extension Not Working">
    **Common Issues**:

    * Restart VSCode after configuration changes
    * Check API key is correctly set in settings
    * Verify internet connection and firewall settings

    <Tip>
      Check the VSCode Developer Console (`Help` → `Toggle Developer Tools`) for error messages.
    </Tip>
  </Accordion>

  <Accordion title="API Rate Limits">
    **Solutions**:

    * Monitor usage in Comput3 dashboard
    * Implement request queuing
    * Consider upgrading your plan

    <Info>
      Rate limits reset every minute. Check your current limits at [app.comput3.ai](https://app.comput3.ai).
    </Info>
  </Accordion>

  <Accordion title="Model Not Responding">
    **Checks**:

    * Verify model name matches available models
    * Test API key with curl:

    ```bash theme={null}
    curl -H "Authorization: Bearer YOUR_API_KEY" https://api.comput3.ai/v1/models
    ```

    * Check Comput3 service status
  </Accordion>
</AccordionGroup>

## Advanced Configuration

### Custom Keybindings

Add to your `keybindings.json`:

```json theme={null}
[
  {
    "key": "cmd+shift+a",
    "command": "codegpt.ask",
    "when": "editorTextFocus"
  },
  {
    "key": "cmd+shift+r", 
    "command": "codegpt.review",
    "when": "editorHasSelection"
  }
]
```

### Workspace Settings

Create `.vscode/settings.json` in your project:

```json theme={null}
{
  "codegpt.apiKey": "${env:COMPUT3_API_KEY}",
  "codegpt.apiBase": "https://api.comput3.ai/v1",
  "codegpt.model": "hermes4:70b",
  "codegpt.temperature": 0.2,
  "codegpt.maxTokens": 2048
}
```

## Benefits

<CardGroup cols={2}>
  <Card title="Native Integration" icon="puzzle-piece">
    Seamless integration with VSCode's existing workflow and shortcuts.
  </Card>

  <Card title="Offline Capable" icon="wifi-slash">
    Continue working with cached responses when connectivity is limited.
  </Card>

  <Card title="Team Consistency" icon="users">
    Shared configuration ensures consistent AI assistance across your team.
  </Card>

  <Card title="Privacy Control" icon="shield-alt">
    Your code stays secure within the Comput3 network infrastructure.
  </Card>
</CardGroup>