Gemini API is constantly becoming overloaded on many models. Throwing GeminiException 503 again.

[Jamie123234]

Gemini API is constantly becoming overloaded on many models. Is it not possible to implement a backup model when this is overloaded? Maybe using llama? I see it seems to be on Googles end, but is there no way around it?

Error in gemini function: **GeminiException** => This exception was thrown because the response has a status code of 503 and RequestOptions.validateStatus was configured to throw for this status code.

The status code of 503 has the following meaning: "Server error - the server failed to fulfil an apparently valid request"

Read more about status codes at https://developer.mozilla.org/en-US/docs/Web/HTTP/Status

In order to resolve this exception you typically have either to verify and fix your request code or you have to fix the server code.

[Jamie123234]

I know @babakcode tried to fix it by retrieving the models dynamically, but it seems that the models on the Gemini API itself are becoming overloaded. Here is a forum article: https://discuss.ai.google.dev/t/error-the-model-is-overloaded/48410