Skip to content

Request for access to ClawsBench evaluation code #1

Description

@rufengdai

Hi ClawsBench team,

I hope this message finds you well. I'm reaching out to request access to the evaluation code for ClawsBench, as I'm working on AI model performance benchmarking and aim to locally reproduce the evaluation process of this benchmark.

I noticed two key points:

  1. The HuggingFace dataset page (https://huggingface.co/datasets/benchflow/ClawsBench) mentions that "open-source task definitions and Dockerized evaluation environment are planned to be released";
  2. The official GitHub repo (https://github.com/benchflow-ai/ClawsBench) currently lacks executable evaluation scripts/code to run local assessments.

Could you kindly share the preliminary version of the ClawsBench evaluation code (even a draft version) or inform the expected timeline for the official release of the open-source task definitions and Dockerized environment? This will help me complete the performance benchmarking of AI models more accurately and reproducibly.

Thank you very much for your hard work on ClawsBench, and I appreciate your time and reply!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions