Gemini Provider

Google Gemini AI models integration addon for AI Core.

Features

• Google Gemini Integration - Full integration with Google's Gemini AI models
• Model Selection - Choose from available Gemini models (Pro, Ultra, etc.)
• API Configuration - Easy setup with Google Cloud credentials
• Multi-Modal Support - Support for text, image, and document inputs
• Streaming Responses - Real-time streaming for long-form content

Requirements

Requirement	Details
Dependencies	AICore
PHP Version	8.2+
Google Cloud	Valid Google Cloud account with Gemini API access

Installation

1. Ensure AI Core module is installed and enabled
2. Enable the Gemini Provider module in Settings > Modules
3. Configure your Google Cloud API credentials

Configuration

Navigate to Settings > AI Core > Providers > Gemini to configure:

• API Key - Your Google Cloud API key with Gemini access
• Default Model - Select default Gemini model
• Safety Settings - Configure content safety thresholds
• Region - Select preferred Google Cloud region

Usage

Using Gemini as Provider

use Modules\AICore\Services\AIService;

$aiService = app(AIService::class);

$response = $aiService->complete([
    'provider' => 'gemini',
    'model' => 'gemini-3-flash',
    'prompt' => 'Analyze this business report...',
]);

Multi-Modal Request

$response = $aiService->complete([
    'provider' => 'gemini',
    'model' => 'gemini-3-pro',
    'prompt' => 'Describe this image',
    'images' => [$imageBase64],
]);

Supported Models

Gemini 3 Series (Latest)

Model	Context	Description
gemini-3-pro	1M tokens	Most intelligent model with frontier intelligence
gemini-3-flash	1M tokens	Default model with fast performance and thinking
gemini-3-deep-think	1M tokens	Advanced reasoning for complex problems

Gemini 2.5 Series (Stable)

Model	Context	Description
gemini-2.5-pro	1M tokens	Superior reasoning with thinking mode
gemini-2.5-flash	1M tokens	Price-performance optimized
gemini-2.5-flash-lite	1M tokens	Most cost-efficient option

Gemini 2.0 Series

Model	Context	Description
gemini-2.0-flash	1M tokens	Fast multimodal model

Notes

• Requires valid Google Cloud billing account
• Usage is tracked through AI Core's usage monitoring
• Supports all Gemini API features including function calling
• Automatic retry with exponential backoff on rate limits