Is there any clear / centralized documentation around multi-modal support for Vertex vs Gemini APIs / Can someone clarify

Gemini API Documentation
1

My current understanding is

Vertex can support

  • gcs resources
  • inline bytes

Gemini can support

  • file Manager references
  • gcs resources
  • inline bytes

Neither supports public / other (i.e. S3) resources

  1. Is this correct?
  2. Is there any documentation / good place to look for staying on top of this?

Thanks!

1 Like
3

Hi @mark_weiss,

Yes, your understanding looks correct based on the current capabilities of Vertex and Gemini APIs.

You can refer these two documentations:

Vertex AI Multimodal Support – Covers supported input types, session configuration, and model capabilities

Gemini API Docs – Explains differences between Gemini Developer API and Vertex AI Gemini API