The dominant technology in hard rock mining is drilling and blasting; therefore, the importance of fragment size estimation is an essential problem in the mining industry. By using a theoretical foundation that explains rock-fracturing mechanisms by blasting, a 3D fragmentation model with general applicability was developed. The main capabilities of the model are the ability to consider different and complex blasting patterns, both with parallel and non-parallel boreholes, and different explosive and rock properties. The geological input of the model is defined by generating the primary blocks within rock mass using discrete fracture networks. The model handles different, complex geological setups and blasting scenarios that include bench, ring, and tunneling blasting. Also, it considers borehole deviation, misfires, and the influence of previous blasts. Full-scale ring blasts were carried out to validate the model against real-world data, and the results showed a high level of agreement between model predictions and in situ data. In situ data were collected using image processing methodology and WipFrag v4 software. X50 and X80 size values showed a maximum error in prediction of around 15%, while R2 values between fragmentation curves were approximately 90% without a detailed model calibration to the obtained data.