LLMaker is a chat-only tool facilitating iterative game content co-design. It serves as a bridge between human designersand a content generation system, utilising LLMs to translate designer requests into instructions for a content generation system. Unlike traditional tools, LLMaker enables designers to express creativity freely via natural language, while ensuring the content adheres to domain rules. LLMaker uses OpenAI’s GPT-3.5 for handling user requests and level design tasks, while a fine-tuned Stable Diffusion model handles visual asset creation.
Roberto Gallotta, Antonios Liapis, Georgios N. Yannakakis: "LLMaker: A Game Level Design Interface Using (Only) Natural Language," in Proceedings of the IEEE Conference on Games, 2024. PDF BibTex
Roberto Gallotta, Antonios Liapis, Georgios N. Yannakakis: "Consistent Game Content Creation via Function Calling for Large Language Models," in Proceedings of the IEEE Conference on Games, 2024. PDF BibTex
Roberto Gallotta, Antonios Liapis, Georgios N. Yannakakis: "The Importance of Context in Image Generation: A Case Study for Video Game Sprites," in Proceedings of the International Conference on Computational Intelligence in Music, Sound, Art and Design (EvoMusArt), 2025. PDF BibTex