Molecular discovery has brought great benefits to the chemical industry. Various molecule design techniques are developed to identify molecules with desirable properties. Traditional optimization methods, such as genetic algorithms,continue to achieve state-of-the-art results across multiple molecular designbenchmarks. However, these techniques rely solely on random walk exploration, which hinders both the quality of the final solution and the convergencespeed. To address this limitation, we propose a novel approach called Gradient Genetic Algorithm (Gradient GA), which incorporates gradient information from the objective function into genetic algorithms. Instead of random exploration, each proposed sample iteratively progresses toward an optimal solution by following the gradient direction. We achieve this by designing a differentiable objective function parameterized by a neural network and utilizingthe Discrete Langevin Proposal to enable gradient guidance in discrete molecular spaces. Experimental results demonstrate that our method significantlyimproves both convergence speed and solution quality, outperforming cuttingedge techniques. For example, it achieves up to a 25% improvement in the top10 score over the vanilla genetic algorithm. The code is publicly available athttps://anonymous.4open.science/r/GradientGA-DC45.