README

connector-gemini

Google Gemini AI connector for Ibexa DXP.

This bundle integrates Google Gemini API with the Ibexa AI framework (ibexa/connector-ai), providing action handlers for text-to-text and image-to-text AI operations.

Requirements

• PHP >= 8.3
• Ibexa DXP ~5.0
• ibexa/connector-ai ~5.0
• Google Gemini API key

Installation

Require the package:

composer require arfaram/connector-gemini

Register the bundle in config/bundles.php:

return [
    // ...
    ConnectorGeminiBundle\ConnectorGeminiBundle::class => ['all' => true],
];

Environment Variable

Set your Gemini API key in .env:

GEMINI_API_KEY=your-google-gemini-api-key

See: https://aistudio.google.com/api-keys

Configuration

Available Gemini Models (latest 4)

Basic Configuration

Create config/packages/ibexa_connector_gemini.yaml:

ibexa_connector_gemini:
    text_to_text:
        default_model: gemini-2.5-flash
        default_temperature: 1.0
        default_max_tokens: 8192
        models:
            gemini-2.5-pro: 'Gemini 2.5 Pro'
            gemini-2.5-flash: 'Gemini 2.5 Flash'
            gemini-2.0-flash: 'Gemini 2.0 Flash'
            gemini-2.0-flash-lite: 'Gemini 2.0 Flash Lite'
    image_to_text:
        default_model: gemini-2.5-flash
        default_temperature: 1.0
        default_max_tokens: 8192
        models:
            gemini-2.5-pro: 'Gemini 2.5 Pro'
            gemini-2.5-flash: 'Gemini 2.5 Flash'
            gemini-2.0-flash: 'Gemini 2.0 Flash'
            gemini-2.0-flash-lite: 'Gemini 2.0 Flash Lite'

SiteAccess-Aware API Key Configuration

You can configure the API key per SiteAccess in config/packages/ibexa.yaml:

ibexa:
    system:
        default:
            connector_gemini:
                gemini:
                    api_key: '%env(GEMINI_API_KEY)%'
        my_siteaccess:
            connector_gemini:
                gemini:
                    api_key: '%env(GEMINI_API_KEY_MY_SITEACCESS)%'

Minimal Configuration (uses defaults)

If you only want to override the default model:

ibexa_connector_gemini:
    text_to_text:
        default_model: gemini-2.5-pro

All other values will use their defaults:

• default_temperature: 1.0
• default_max_tokens: 8192
• models: all 4 latest Gemini models

Production-Optimized Configuration

For production environments favoring speed and cost:

ibexa_connector_gemini:
    text_to_text:
        default_model: gemini-2.5-flash
        default_temperature: 0.7
        default_max_tokens: 4096
        models:
            gemini-2.5-flash: 'Gemini 2.5 Flash'
            gemini-2.0-flash-lite: 'Gemini 2.0 Flash Lite'
    image_to_text:
        default_model: gemini-2.5-flash
        default_temperature: 0.5
        default_max_tokens: 2048
        models:
            gemini-2.5-flash: 'Gemini 2.5 Flash'
            gemini-2.0-flash: 'Gemini 2.0 Flash'

License

MIT - See LICENSE for details.

Author

Ramzi Arfaoui - ramzi_arfa@hotmail.de