This research demonstrates how fine-tuned LLMs can automatically generate executable chemical reactor models from natural language descriptions.

• Fine-tuned Llama 3.1 8B to produce Modelica code for various chemical reactor scenarios
• Converts simple text descriptions into functional reactor models
• Streamlines complex engineering tasks through natural language interfaces
• Represents a practical application of AI for chemical engineering workflows

Why it matters: This approach could significantly accelerate reactor design processes, reduce manual coding requirements, and make advanced modeling more accessible to engineers without extensive programming experience.

Text2Model: Generating dynamic chemical reactor models using large language models (LLMs)