Games
SubwayBench
A one-shot benchmark comparing frontier models on complex 3D game generation.
Subway Surfers cloneOpen source
Description
A technical benchmark that evaluates how 12+ frontier AI models handle a complex prompt to build a 3D Subway Surfers clone in a single shot. The analysis identifies Claude Opus 5 as a category-leading winner for implementing a true projected 3D engine, frame-rate independent physics, and advanced platformer affordances like coyote time and jump buffering from a single prompt.
Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.