Support for metadata labels in VertexAI library

[btrajkovski-mms]

Thanks for stopping by to let us know something could be better!

PLEASE READ: If you have a support contract with Google, please create an issue in the support console instead of filing on GitHub. This will ensure a timely response.

Is your feature request related to a problem? Please describe. Java library for VertexAI does not offer support for adding labels when making generateContent calls. This is really useful feature for tracking costs for different labels. And also this is supported by REST call and by python library. Link - https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/add-labels-to-api-calls

Describe the solution you'd like Solution could be the same as in python library

GenerativeModel::generateContent(Content content, Map<String, String> labels)

This is the example from python library:

prompt = "What is Generative AI?"
response = model.generate_content(
    prompt,
    # Example Labels
    labels={
        "team": "research",
        "component": "frontend",
        "environment": "production",
    },
)

Describe alternatives you've considered Unfortunately no alternatives in this library. If I want to use labels I need to switch to REST calls and discard this library.

[diegomarquezp]

cc: @jaycee-li @sasha-gitg

[twistedpair]

This would really help when scaling AI models rollout, so you can segment and understand your AI cost centers. Otherwise you need to log token counts and labels yourself, push to BQ, do analytic rollups, etc. Labels are a common feature across GCP and its SDKs.