Calling gemini1.5 pro API, using Stream mode, continuous error occurs, what is the reason
Hi @picmagic_Nick, Welcome to the forum!!
What error you are getting? Can you help with providing the code snippet if possible to repro the issue ?
Thanks.
Hi Govind_Keshari:
I use LiteLLm SDK to call “gemini/gemini-1.5-pro”, below is the error message:
litellm.exceptions.InternalServerError: litellm.InternalServerError: litellm.InternalServerError: VertexAIException - b’{
“error”: {
“code”: 503,
“message”: “The model is overloaded. Please try again later.”,
“status”: “UNAVAILABLE”
}
}
httpx.HTTPStatusError: Server error ‘503 Service Unavailable’ for url ‘https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-pro:streamGenerateContent?key=xxxxxxx&alt=sse’
Model overload is often temporary. Wait a few minutes, or even a few hours, and try your request again. The load on the model might have decreased.
If it was working earlier and now you are getting the error then @KRows is right. Retry your request.