In order to reduce crack occurrence during the laser additive manufacturing of TiAl alloy, a predictive model was developed to analyze the relationships between process parameters (laser power, scanning speed, and powder feed rate) and crack formation. The study utilized the number of cracks as the response index to investigate the influence of process parameters and their interactions. Response surface and genetic algorithm techniques were used to optimize the process parameters, and the microstructure near the crack and the root cause of crack formation were examined. Results underscored the significant impact of laser power on crack occurrence, with fewer cracks observed at higher laser powers. Comparative analysis of optimization algorithms favored the genetic algorithm, achieving crack free samples with optimal parameters: laser power of 1 754 W, scanning speed of 9mm/s, and powder feed rate of 7g/min. Furthermore, the volatilization of Al near cracks facilitated the formation of the brittle α2 phase, while localized enrichment of α2 phase rendered cracks more susceptible to stress induced failure during the laser additive process.