Basic question: what is the difference between the Gemini API and Vertex?

Gemini API
1

The API is working for me fine as is, why would I want to go to Vertex?

2

Vertex works in more countries, some people are already experienced with Google Cloud, so it’s easier for them to use it. The UI is easier to use and has more functionality. Vertex supports PDFs.

1 Like
3

If the Gemini API through the AI Studio API works well for you - great! No need to switch, of course.

The Vertex AI offers a number of things that people expect and need from a cloud platform, including easy integration with other components of the Google Cloud Platform. Some of the implications of this are:

  • Data residency. Some countries require you to keep customer data in a specific region. Vertex AI lets you pick which data center Gemini runs in for you, while AI Studio makes no specific guarantee.
  • Are you going to run your service on something like Google Cloud Functions? You can take advantage of Application Default Credentials and not have to code or store an API Key or Service Account anywhere. Your code automatically runs as a service account, and you can just permit that to access the Gemini API. No Keys or credentials to setup.
  • Are you storing your images, videos, and audio in Google Cloud Storage already? You can provide the gs: URI to Gemini when referencing fileData instead of having to use the AI Studio File API
  • Vertex AI offers tools that let you ground your prompt against other services on Google Cloud that AI Studio doesn’t
    • Grounding against data in Google Search
    • Grounding against data in a Vertex AI Search Data Store, which can get it’s data from documents in GCS, data in BigQuery, or a variety of other sources.
  • Quota, billing, and support services. Although AI Studio does billing through Google Cloud, there are still big issues with quota and quota requests. Vertex AI handles this through long-established system.
1 Like