THUSI-Lab / GameScaling

Scaling in Games: Continued Pre-Training for Embodied Agents in Diverse Virtual Worlds. Weights on Hugging Face.

kikoncuo / jevfire

JEV-inspired parallel decisions for CUDA LLMs. One context, many decisions. vLLM API, game-agent examples, and reproducible benchmarks.

guangjiaohuoqiubumingzhong / Nagi--galgame-agent-

—

Eurekaleo / awesome-ai-for-games

A curated, searchable collection of research on AI and foundation models across the game lifecycle.