Google Gemini API Models does not work with OpenAI compatibility

[akhilesh-d]

Bug Report

Steps to Reproduce

1. Add model in GPT4ALL with OpenAI-compatible options.
2. Provide BASE_URL as https://generativelanguage.googleapis.com/v1beta/openai
3. BASE MODEL as gemini-1.5-flash and a valid API KEY e.g. Azqqdfwerregb
4. Use this model to send a question in chat
5. No response is printed while same API call work via REST API using CURL

Expected Behavior

Your Environment

• GPT4All version: 3.8.0
• Operating System: Windows 11
• Chat model used (if applicable): gemini-1.5-flash

[isaiah2004]

can confirm the same is true with gemini 2.0 models it's for some reason not working.

[akhilesh-d]

after upgrade to GPT4all 3.10

ERROR: ChatAPIWorker::handleReadyRead got HTTP Error 400 Unknown error: [{
"error": {
"code": 400,
"message": "Unable to submit request because it has an empty text parameter. Add a value to the parameter and try again. Learn more: https://cloud.google.com/vertex-ai/generative-ai/docs/model-reference/gemini",
"status": "INVALID_ARGUMENT"
}
}
]

[akhilesh-d]