Would you add support for other models as well

[dinuransika]

Description

Will you add support for models like VLA0 in the future.

[MilkClouds]

Thanks for the interest! Yes — adding new model servers is straightforward by design. The harness decouples model serving from benchmark execution, so integrating a new model is mostly a single Python file + a YAML config. See CONTRIBUTING.md for the step-by-step guide.

We'd welcome a community PR for VLA0 or any other model. As a reference, each model server is a ~100-line PredictModelServer subclass (e.g., pi0.py). If you'd like to give it a try, feel free to open a PR — happy to help with review.