README

Gemini provider for adachsoft/ai-integration using the official OpenAI-compatible chat completions endpoint exposed by Google.

This package exposes a ready-to-use SPI provider that can be registered under the id gemini-via-openai in the core ToolCallingChatFacadeBuilder and used for tool-calling chat scenarios.

Installation

composer require adachsoft/ai-integration adachsoft/ai-integration-gemini-via-openai

Configuration

Set the following environment variable (for example in your .env file at the project root):

• GOOGLE_GEMINI_API_KEY – API key for the Gemini OpenAI-compatible endpoint.

Usage

use AdachSoft\AiIntegration\PublicApi\Builder\ToolCallingChatFacadeBuilder;
use AdachSoft\AiIntegration\PublicApi\ToolCalling\Dto\ChatMessageDto;
use AdachSoft\AiIntegration\PublicApi\ToolCalling\Dto\Collection\ChatMessageDtoCollection;
use AdachSoft\AiIntegration\PublicApi\ToolCalling\Dto\Collection\ToolDefinitionDtoCollection;
use AdachSoft\AiIntegration\PublicApi\ToolCalling\Dto\ToolCallingChatRequestDto;
use AdachSoft\AiIntegration\PublicApi\ToolCalling\Dto\ToolDefinitionDto;
use AdachSoft\AiIntegrationGeminiViaOpenAi\GeminiViaOpenAiToolCallingChatSpiFactory;

$apiKey = getenv('GOOGLE_GEMINI_API_KEY');

$spi = GeminiViaOpenAiToolCallingChatSpiFactory::create($apiKey);

$facade = ToolCallingChatFacadeBuilder::create()
    ->withSpiProvider('gemini-via-openai', $spi)
    ->build();

$tools = new ToolDefinitionDtoCollection([
    new ToolDefinitionDto(
        name: 'echo_tool',
        description: 'Echoes back the provided input string.',
        parametersSchema: [
            'type' => 'object',
            'properties' => [
                'text' => [
                    'type' => 'string',
                    'description' => 'Text to echo back',
                ],
            ],
            'required' => ['text'],
        ],
    ),
]);

$messages = new ChatMessageDtoCollection([
    ChatMessageDto::createUserMessage('Call echo_tool with some example text.'),
]);

$request = new ToolCallingChatRequestDto(
    providerId: 'gemini-via-openai',
    modelId: 'gemini-2.0-flash',
    messages: $messages,
    tools: $tools,
);

$response = $facade->chat($request);

Smoke test

A simple smoke script is provided for manual end-to-end verification:

echo "GOOGLE_GEMINI_API_KEY=your_api_key_here" >> .env
php bin/smoke-gemini-via-openai-tool-chat

The script will perform a single tool-calling chat request using the gemini-2.0-flash model and print the response in JSON form.