This topic is a continuation of the following article.
https://www.googlecloudcommunity.com/gc/AI-ML/Setup-API-gateway-to-call-vertex-ai-endpoint/m-p/63390...
I want to use the Gemini API and Python's Vertex AI SDK with user-specific authentication information, while logging on the GCP side.
(And I would like the logging to be recorded on the infrastructure side, rather than being explicitly set by the user client.)
Is there still no other way but to use Google Cloud Functions for mediation in this case?
Are there any plans to enable direct calls to the VertexAI API (GeminiAPI) from API Gateway?
Currently, when writing OpenAPI specifications, the GeminiAPI includes a colon in its URL, making it difficult to properly write OpenAPI and configure API Gateway (such as "gemini-1.5-pro-001:streamGenerateContent").
It would be greatly appreciated if you could provide any sample code or examples on how to execute VertexAI through Google Cloud Functions and API Gateway.
LinkedIn
Twitter
