DishwasherBench – Can Frontier Models Stack a Dishwasher?

A GitHub repository called DishwasherBench has been created to investigate whether Frontier models can be stacked to achieve the same level of performance as larger models. The project aims to explore the possibility of stacking multiple smaller models to replicate the capabilities of larger models. The repository contains a script to run experiments and a notebook with instructions on how to use it. The project's results are not yet available, and it is unclear what kind of performance Frontier models can achieve when stacked. The project is still in its early stages, and the community is invited to contribute to the research and experiments.