Skip to content

Add ensemble_batch_size for single-device inference #905

@randommm

Description

@randommm

Describe the workflow you want to enable

Add ensemble_batch_size for single-device inference

Describe your proposed solution

On devices with large amounts of RAM like Strix Halo, this can greatly speed up results

Describe alternatives you've considered, if relevant

No response

Additional context

No response

Impact

None

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