The Network Center of the Institute of Physics, Chinese Academy of Sciences released the AI model MatChat to predict the synthesis path of inorganic materials.

CTOnews.com, Nov. 3, the official website of the Institute of Physics of the Chinese Academy of Sciences posted that recently, the Institute of Physics of the Chinese Academy of Sciences / SF10 Group of the Beijing National Research Center for condensed matter Physics and the computer Network Information Center of the Chinese Academy of Sciences have worked together to apply the AI model to the field of material science, feeding tens of thousands of chemical synthesis path data to the large language model LLAMA2-7b, thus obtaining the MatChat model. It can be used to predict the synthesis path of inorganic materials.

CTOnews.com noted that the model can be logically inferred according to the asked structure and output the corresponding preparation process and formula. At present, it has been deployed online and is open to all material researchers, which brings new inspiration and new ideas for material research and innovation.

This work puts forward a possible solution for the application of large language model in the field of subdivision science, and preliminarily shows the feasibility of the method. Extracting literature data through natural language model, and then applying literature data to language model training is a feasible way to develop subdivision science field. This work shows the "one-minute flight of the Wright brothers" in the field of synthetic path prediction of inorganic materials, and the prediction accuracy of the model is still limited by the quantity and quality of data sets.