Skip to content

configuration.md

Latest commit

 

History

History
29 lines (21 loc) · 1.21 KB

configuration.md

File metadata and controls

29 lines (21 loc) · 1.21 KB
Raw

Configuration

Configuring Gemini PHP is straightforward. You need to provide several key pieces of information that the library will use to authenticate and interact with the Gemini AI platform.

Configuration Parameters

  • projectId: Your Google Cloud Project ID.
  • regionName: The Google Cloud region where your Vertex AI API is hosted.
  • credentialsPath: The file path to your service account credentials JSON.
  • modelName: The name of the AI model you wish to use, such as gemini-pro or gemini-pro-vision.
  • ignoreModelValidation: Set to true to not be constricted to gemini-pro or gemini-pro-vision.
  • ignoreRegionValidation: Set to true to not be constricted to hardcoded regions.

Example Configuration

Here's an example of how to configure the GeminiClient:

$config = [
    'projectId' => 'ai-project-123456',
    'regionName' => 'us-east4',
    'credentialsPath' => 'path-to-your-credentials.json',
    'modelName' => 'gemini-pro',
];

$client = new GeminiClient($config);

Ensure that the credentialsPath points to a valid JSON file containing your service account credentials. The modelName should correspond to the specific AI model you intend to use for generating content.