GAMEGO: Training Game-Dev Agents with Synthetic Trajectories Anchored in Real-World Assets

A new arXiv paper (2610.06910v1) presents GameGo, a framework that transforms brief game seeds into comprehensive Product Requirements Documents and was used to build GameGoData, a dataset of 55,060 development trajectories spanning 2D, 2.5D, and 3D games, plus GameGoBench, a benchmark of 124 game queries. Training GameGoCoder on GameGoData produced a model that outperforms matched baselines and is comparable to frontier models across game-development benchmarks, with all code, datasets, and models to be made publicly available.

arXiv:2610.06910v1 Announce Type: new Abstract: Recent advances in Large Language Models LLMs have demonstrated remarkable capabilities in web front-end execution, with browser-based game generation emerging as a particularly prominent frontier. While previous efforts frequently rely on complex multi-turn workflows or focus on static game evaluation benchmarks, this work targets direct end-to-end real-world game synthesis driven by coding agents. However, generating complex games directly from sparse user queries often forces coding agents to make underspecified assumptions, yielding incomplete mechanics, disconnected gameplay flows, and limited visual aesthetics. To resolve this issue, this paper presents GameGo, a scalable framework that systematically transforms brief game seeds into comprehensive Product Requirements Documents grounded in industry game-development practices. To retain core gameplay constraints without restricting design exploration, GameGo uses task-specific dynamic compression to maximize information density while preserving instruction following. Based on this pipeline, GameGoData is constructed with 55,060 development trajectories across 2D, 2.5D, and 3D games, alongside GameGoBench, a benchmark comprising 124 diverse game queries. Training GameGoCoder on GameGoData yields a model that outperforms matched baselines and is comparable to frontier models across gamedev benchmarks. All code, datasets, and models will be made publicly available.