Benchmark Radar™
Star —

GameCraft-Bench

Explore score history ↗

A benchmark of 140 Godot engine tasks across 15 game families. Coding agents must build playable games end-to-end, and the games are scored through replayed demonstrations and rubric-guided multimodal judging.

Released
2026-06-16
Institutions
Chinese University of Hong Kong, Shenzhen · Tencent (China) · National University of Singapore
Evaluates
General AI, Language & Knowledge, cs.CL
Openness
unknown
Importer
Claire Radar
Review status
ai-reviewed

GameCraft-Bench paper

No reported scores are on record for this benchmark yet.

Which sources cite GameCraft-Bench?

Related General AI, Language & Knowledge, cs.CL benchmarks