Skip to content

Support other VLMs for object detection #15

@williamshen-nz

Description

@williamshen-nz

Description

Right now we use frontier Gemini model for detection and goal translation. Might be achievable with Gemma 4, Molmo, etc.

Motivation

No need for Gemini API key. Easier to experiment with local models.

Alternatives Considered

No response

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions