Error 404 when using gemini-1.5-flash with generateContent

[Trsistan]

Hello,

I encountered an issue while using the Gemini API. When I try to call the generateContent endpoint with the model gemini-1.5-flash, I get the following error:

[GoogleGenerativeAI Error]: Error fetching from https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-flash:generateContent:
[404] models/gemini-1.5-flash is not found for API version v1beta, or is not supported for generateContent. Call ListModels to see the list of available models and their supported methods.

It seems that gemini-1.5-flash is no longer available for this endpoint in the v1beta API version. I have verified that the base URL and other API calls are working correctly.

Could you please confirm if this model has been retired? If so, it would be very helpful to update the documentation to indicate which models are currently supported, for example gemini-2.5-flash-lite.

Thank you very much for your help and for maintaining this API!

Best regards,

[rodolfo-terriquez]

I think the model has been deprecated by google - https://cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versions#retired-models

[Marca64]

Yes, that's why we need an update adding the newer models (or letting users specify the models themselves).

[albortino]

Quick fix:

1. Locate the obsidian folder.
2. Go to .obsidian/plugins/quiz-generator/.
3. Update the data.json file. Instead of gemini-1.5*, write gemini-2.5*. This sets the default model for inference. However, be aware that once you overwrite it within the plugin settings, it will be lost.
4. If you also want to persist changes and be able to select the models from a dropdown menu, change it in the main.js file as well. Be cautious, as it is not well-formatted.

To make the changes, simply open the file and replace the string gemini 1.5 with gemini 2.5 and gemini-1.5 with gemini-2.5.

As of my last update, the current models are the Gemini 3 models (e.g., gemini-3-flash-preview). Of course, you can also change the strings to those models, but I consistently encounter an „model is overloaded“ error when using the 3-preview models, while the 2.5 models work fine.

Note: When updating the quiz-generator plugin, these changes may be lost. However, we hope that this issue will be resolved in the next update.