Show HN: Pac-Bench – How well can models one-shot a Pac-Man game?

Pac-Bench is a benchmark that tests how well large language models can create a functional Pac-Man game from a single prompt. The models, including Harness, are given a single chance to create the game in HTML without any follow-up prompts or fixes. The benchmark aims to evaluate the capabilities of these models in generating complex software from simple text-based input.