In order to scientifically and reasonably evaluate the ecological efficiency of low-carbon enterprises, this article proposes a neural network-based ecological benefit evaluation strategy for low-carbon enterprises. Firstly, the concept of low-carbon economy is introduced, and the evaluation indicators for low-carbon supply chain performance of steel enterprises are established from three aspects of economy, energy, and environment for filtering, integrated with the connotation of energy conservation and emission reduction to establish an indicator system for low-carbon supply chain performance evaluation of steel enterprises. Then, a BP neural network model is constructed for simulation testing, and neural network is adopted to train and test these sample data to calculate the comprehensive scores of various ecological benefit indicators. Finally, through empirical analysis of the problems in the low-carbon supply chain in the case, it is proved that the proposed solution has good adaptability and self-regulation ability, and it can more accurately reflect the low-carbon efficiency level of the enterprise.