Apparel pattern modification guidance of competing LLMs: ChatGPT, Claude and Gemini

Purpose This study seeks to benchmark the current capabilities of large language models (LLMs) to diagnose and address fit issues in garment pattern making. Design/methodology/approach Five garments were created using three-dimensional garment models that exemplified common fit issues. This visual input was used as a prompt for artificial intelligence (AI) tools such as ChatGPT, Claude and Gemini. The AI output was then evaluated by industry experts to assess their accuracy. Findings The LLMs displayed varying success, with Claude performing better than other GPTs in some tests. None of the AIs consistently provided accurate results, however, underscoring the need for further development in AI patternmaking assistance. Originality/value The possibility of customizing LLMs with domain-specific source material shows some promise in elevating accuracy. The study illustrates the limitations of current LLMs and highlights the potential for future use of AI in educational and professional settings, particularly in fields like patternmaking where acquired skill and tacit knowledge are required.