ChemMLLM: Chemical Multimodal Large Language Model

TL;DR

ChemMLLM is a novel multimodal large language model tailored for chemical molecule understanding and generation, integrating text, SMILES, and images, and demonstrating superior performance on curated chemical tasks.

Contribution

The paper introduces ChemMLLM, the first unified chemical multimodal LLM, with designed tasks and datasets, advancing cross-modal understanding and generation in chemistry.

Findings

ChemMLLM outperforms baseline models on all tasks.

Achieves 116.75% improvement in molecule image optimization.

Code is publicly available for reproducibility.

Abstract

Multimodal large language models (MLLMs) have made impressive progress in many applications in recent years. However, chemical MLLMs that can handle cross-modal understanding and generation remain underexplored. To fill this gap, we propose ChemMLLM, a unified chemical multimodal large language model for molecule understanding and generation. Also, we design five multimodal tasks across text, molecular SMILES strings, and image, and curate the datasets. We benchmark ChemMLLM against a range of general leading MLLMs and Chemical LLMs on these tasks. Experimental results show that ChemMLLM achieves superior performance across all evaluated tasks. For example, in molecule image optimization task, ChemMLLM outperforms the best baseline (GPT-4o) by 116.75\% (4.27 vs 1.97 property improvement). The code is publicly available at https://github.com/bbsbz/ChemMLLM.git.