In studying connectome organization, generative network models (GNM) have become a critical tool in describing the underlying resource and network constraints driving development. While GNMs highlight nuanced economic trade-offs, they exclusively describe binary networks, without regard for the strength of connections. Weighted generative models (WGM), provide a promising expansion to the current strategy. By leveraging communicability and redundancy calculations, WGMs can modulate edge values within a network throughout development and successfully capture the topology of a biological weighted network. This study aims to test the validity of the WGM in predicting proper connectome development and features of a network at the cellular scale using neural data. We fit our model to spike time tiling coefficient (STTC) calculated networks from Microelectrode array (MEA) recordings of developing rodent cortical cultures and various organoid slices. Parameters from the WGM equations were selected across a search window and then used to simulate a network. Fit was assessed by the similarity of topological properties such as clustering, betweenness, strength, and edge length between the simulated and empirical connectomes. We successfully fit the WGM to data at the cellular scale and observed topological trends from the biological networks accurately captured by the model. Further, we noted fitted parameter values that imply more stochastic development in plated cells as compared to past whole-brain studies. Our study validates this preliminary model’s ability to analyse differences between neuronal recordings, evidencing the driving factors of development and making a strong case for future WGM experiments.