Skip to content

Normalize runtime timings by the number of tokens #41

@rlouf

Description

@rlouf

The total runtime is not interesting, what we care about is the latency per token generated. We should therefore report this number instead of the total runtime.

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions