client icon indicating copy to clipboard operation
client copied to clipboard

Google Gemini API PHP Client allows you to use the Gemini AI model

Results 32 client issues
Sort by recently updated
recently updated
newest added

Hi, I don't see anything about function calling, is it available in the sdk ? Thanks

Just a suggestion: maybe it would make sense to allow passing a custom string as model name. I wanted to test `models/gemini-1.5-pro-latest` but its not part of the `ModelName` enum....

enhancement

When testing locally I get this error: `PHP Fatal error: Uncaught RuntimeException: Gemini API operation failed: operation=models/gemini-pro:generateContent, status_code=400, response={ "error": { "code": 400, "message": "User location is not supported for...

According to google docs here https://ai.google.dev/docs/gemini_api_overview#json We can instruct Gemini to return `json` by adding `"response_mime_type": "application/json"` to `generationConfig` I hope this will be added to `generationConfig`

Hello, First, thank you for great API PHP Client! This is something Google seems to forget, great support for PHP. While trying your examples, I got error: Fatal error: Uncaught...

I need to use proxy because api doesnt support in my region, where should i fix it ? i couldnt find it? where is curl?

if so, Can you please provide me some sample code for the multimodal video? This is my working code, Multimodal image. `function AiFunction($text, $instructions, $image_urls = []) { global $GOOGLE_AI_CLIENT;...

[Multimodal input](https://github.com/gemini-api-php/client#multimodal-input) doesn't work anymore, what alternative can I use? Does anyone know? I don't have much knowledge. I only know enough to get it running, that's all. 🤦‍♂️