Skip to content

The new version does not provide detailed chunks as the old version does #108

@sangtrx

Description

@sangtrx

the "old version" appears to output finer-grained descriptions and potentially more segmented chunks (e.g., detailed scene breakdowns for figures like the cartoon character with a telescope), while the "new version" consolidates elements into broader summaries (e.g., a high-level overview of the fishbowl illustration)

How could I get the result like the old version?
Old:
Image
New:
Image

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions