Invalid request error in Google Flash

Support
1

Hi all,

I’m currently using Google Flash 2.0 (Gemini 2.0 Flash) with a direct API key from Google and have configured the system to use the Google Tokenizer. The chat functionality is working perfectly; however, when I try to enable image support, I encounter the following error:

{ "error": { "message": "'$.input' is invalid. Please check the API reference: https://platform.openai.com/docs/api-reference.", "type": "invalid_request_error", "param": null, "code": null } }

Has anyone else seen this? Is this error due to Gemini 2.0 not yet supporting vision features in Discourse AI? Any insights or workarounds would be appreciated.

Thanks!

3

Yes I think I do :slight_smile:

You are trying to use the Open AI compatible endpoint Google provide instead of the Gemini native endpoint.

Try using:

https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash

And use the Gemini provider.

The OpenAI compatibility layer has some limitations.

4

Thanks Sam, that is the URL for the setting. The error message occurs when using a persona with that LLM and also set for Vision Enabled and you upload a picture. It fails after hitting upload for jpg. I try to validate as best possible because there are so many settings it’s easy to miss something.

The image shows a settings page for an AI model called Gemini-2.0-flash, provided by Google with an API key input, a "Run test" and "Submit" button. (Captioned by AI)
The image shows a settings page for an AI model called Gemini-2.0-flash, provided by Google with an API key input, a "Run test" and "Submit" button. (Captioned by AI)595×1285 46.3 KB

The image shows a message indicating an error occurred while trying to reply, with detailed error information that explains the input is invalid. (Captioned by AI)
The image shows a message indicating an error occurred while trying to reply, with detailed error information that explains the input is invalid. (Captioned by AI)1107×612 24.3 KB

1 Like
5

It is working here, something feels a bit odd.

The image shows a configuration screen for a Google AI tool named "Gemini 2.0 Flash," detailing its model ID, service URL, API key, and settings. (Captioned by AI)
The image shows a configuration screen for a Google AI tool named "Gemini 2.0 Flash," detailing its model ID, service URL, API key, and settings. (Captioned by AI)815×1304 162 KB

Can you share the output of copy request and response on a gist?

The image displays a configuration screen for a Google AI tool, outlining its service settings, key, model ID "genmin-2.0-flash," service URL, and enabled features like GeminiToKenizer and Vision. (Captioned by AI)
The image displays a configuration screen for a Google AI tool, outlining its service settings, key, model ID "genmin-2.0-flash," service URL, and enabled features like GeminiToKenizer and Vision. (Captioned by AI)989×1187 134 KB

6

Where do I see that screen?

7

That is the i button you have enabled at the bottom.