Is Gemini Pro on Vertex AI down right now?

Google Cloud Build with AI Custom ML & MLOps
,
1

I keep getting either a null response from my requests for generateContent on node, or one of these 2 errors:

ClientError: [VertexAI.ClientError]: got status: 429 Too Many Requests. {“error”:{“code”:429,“message”:“Traffic is exceeding serving capacity, please reduce rate or try again later.”,“status”:“RESOURCE_EXHAUSTED”}}
GoogleGenerativeAIError: [VertexAI.GoogleGenerativeAIError]: got status: 500 Internal Server Error. {“error”:{“code”:500,“message”:“Internal error encountered.”,“status”:“INTERNAL”}}

My quotas are pretty high for generating content so I know I’m not hitting any quotas. Everything was working fine until midway through the day yesterday, and I am using the exact same prompts with the exact same settings today.

Any help is appreciated, thanks!

1 Like
2

Generally its on a quota side, another possible cause that your backend is retrying too quickly, It is recommended that you apply a back off on your requests:

https://cloud.google.com/vertex-ai/docs/general/troubleshooting#error_code_429

3

I am also getting the error, “Error: GoogleGenerativeAIError: [VertexAI.GoogleGenerativeAIError]: got status: 500 Internal Server Error. {“error”:{“code”:500,“message”:“Internal error encountered.”,“status”:“INTERNAL”}}”

(How) can we specify when calling generativeModel.generateContent() to return an error message that is more descriptive of the problem?

1 Like