The original usage seems broken: new GenerativeModel(...)

[doggy8088]

Here is my code:

async Task Main()
{
    var model = new GenerativeModel(apiKey: Util.GetPassword("GEMINI_API_KEY"), model: Model.GeminiPro);

    var prompt = "I love Taipei.";

    var response = await model.GenerateContent(prompt);

    response.Text.Dump();
}

LINQPad Query: https://share.linqpad.net/s5ftv46u.linq

Screenshot:

[jochenkirstaetter]

You got an HTTP 429 Too many requests.

You exceeded the default limit 60 qpm (queries per minute).

[doggy8088]

I know the limitation of the 60 QPM. The point is that, I am NOT request that much at that moment. The Google AI Gemini API will returned HTTP 429 from time to time. Unknown reason! Don't you met that before?

[jochenkirstaetter]

Hardly. However, that's an external issue which I cannot control in the package itself. That's probably something that could be handled by a Retry pattern/approach. There are times when Google AI has too many loads and rejects calls even though you might be below the 60 QPMs. It's discussed among other developers, too.

[jochenkirstaetter]

Considering this as a use case of #2