Abstract: According to the industrial challenge where consumers’ personalized demands are unmet and feedback is inaccurately captured by merchants, this paper introduces an enhanced GPT model based on Kansei engineering, termed “Kansei engineering GPT”, and proposes a distributed customized manufacturing solution based on it. The solution fine-tunes the GPT language model, trains it on a customized dataset, and integrates a sentiment mining approach to generate products that meet user personalization. Digital prototypes are blockchain-uploaded to safeguard intellectual property and sensitive information. Addressing transparency and variability in distributed manufacturing, we introduce a task allocation and pricing mechanism using multi-attribute inverse auction and blockchain.Finally,by establishing a blockchain distributed 3D manufacturing management platform based on Kansei engineering GPT and verifying the feasibility of the task allocation strategy proposed in the paper in a distributed manufacturing environment by comparing it with other allocation methods, it is generally proved that this perceptual engineering GPT-driven distributed customized manufacturing scheme can effectively realize the user's personalized needs in a distributed environment.