Skip to content

[Feature Request]: Enabling 'Parts' for Gemini Requests to Google, to make use of Multi-Modal GCP features #1611

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
mattbeard0out opened this issue Apr 10, 2025 · 2 comments
Labels
enhancement New feature or request

Comments

@mattbeard0out
Copy link

Is your feature request related to a problem? Please describe.

Gemini models hosted on Google Cloud Services offer options of including 'Parts' in messages, for multi-modal interaction. This includes directly linking to Google Buckets, without the need to have the model fetch it themselves, becoming more deterministic.

This is missing from AG2, with no support for files as Parts as part of the framework.

Describe the solution you'd like

For Gemini models, users should be able to use Parts from google types python package, to construct messages. This is important for enterprise could deployments.

Additional context

Image

From Model Garden on Google

@mattbeard0out mattbeard0out added the enhancement New feature or request label Apr 10, 2025
@pasindu-pr
Copy link

@mattbeard0 Thanks for opening this feature request. Google GenAI also provides a separate file upload feature, which you can check out here:https://ai.google.dev/api/files

@mattbeard0
Copy link
Collaborator

@mattbeard0 Thanks for opening this feature request. Google GenAI also provides a separate file upload feature, which you can check out here:https://ai.google.dev/api/files

Thanks, this is good, but might not be aligned with instances where files already exist in GCS.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

3 participants